PREFACE

This paper describes the behaviour of materials
exposed to fire. The performance of precast concrete
and precast reinforced concrete when exposed to stand-
ard fire test is analysed and the influence of the
properties of the aggregates, the cement paste, and the
reinforcing steel on fire endurance of the completed
assembly is discussed. The author Dr. Ing. H. J. Wierig
was formerly engaged in Fire Testing with the University
of Braunschweig and is now with the Cement and Concrete
Research Institute in Beckum, West Germany.
The Division of Building Research extends its thanks
to Mr. D.A. Sinclair of the Translations Section of the
National Research Council who translated this paper and
to Mr. M. Galbreath of this Division who checked the
translation.
Ottawa
April, 1968
R. F. Legget
Director
NATIONAL RESEARCH COUNCIL OF CANADA
Technical Translation 1317
Title: The behaviour of concrete products and precast reinforced
concrete members in a fire
(Das Verhalten von Betonwaren und Stahlbetonfertigteilen im
Feuer)
Author: H.J. Wierig
Reference: Betonstein-Zeitung, 29 (8): (9): (10):
503-510, 1963
(Reprinted in Untersuchungen an Beton, 6: 3-31, 1964)
Translator: D. A. Sinclair, Translations Section, National Science Library
THE BEHAVIOUR OF CONCRETE PRODUCTS AND PRECAST REINFORCED
CONCRETE MEMBERS IN A FIRE
1. Introduction
Fire is one of the natural forces to which man has always been
exposed. From antiquity down to the present, history tells of many cat-
astrophic fires which destroyed whole villages and cities. At an early
date attempts were being made to profit by the experience gained from fire
and to adapt the construction of buildings and to plan villages and cities
accordingly. As an example we may cite the well-known fire of London in
1666. At the time of reconstruction of the city, very extensive building
regulations were enacted(l). For example, the kinds of building materials
to be used were specified, and the width of the new streets was determined.
These regulations can be regarded as the forerunners of our present-day
regulations.
However, systematic research into the behaviour of building materials
and structures exposed to fire was only begun about 100 years ago. At first
the investigations were restricted to the burning down of old buildings
condemned to demolition(1,2). Later, systematic fire tests under well-
defined conditions were introduced. Today, in most countries, there are
well-defined standards and test procedures, and an international standard
is under preparation. In Germany fire tests are carried out under DIN 4102
"Resistance of building materials and members to fire and heat". This
standard is at present under revision. A draft of the testing regulations
contained in the revised version of this standard has just been pUblished.
In the last 100 years a great deal of knowledge has been gained
about the behaviour of building materials and members in fires. Nevertheless,
the damage caused by fire has continued to increase even in recent years. In
Figure 1 the finely hatched columns show the increase in damage due to fire in
the Federal Republic of Germany during the years 1954 to 1961(3). The growth
of the gross national products reduced 1000 times is shown for the same
period by the coarsely hatched cOlumns(4). It will be recognized that the
damage has been increasing at a somewhat faster rate than the gross national
product. Similar trends have been noted in other countries. Figure 1
illustrates very clearly that all economically feasible measures must be taken
in order to prevent any further increase of damage due to fire. In this
connection preventive structural fire protection acquires a special impor-
tance. Structural measures include the following:
-4-
the prevention, as far as possible, of the outbreak of
measures hindering the spread of fire; and provision of escape
routes for the use of the inhabitants and occupants of a building
in the event of a catastrophic fire.
For the prevention of the spread of fire damage the load-bearing
and separating members of a building must resist the fire for a sufficiently
long time, i.e. they must continue to exercise their intended functions.
The required fire resistance time depends on the nature and function of the
building involved. For example, different regulations are applied to ware-
houses and theatres in the building codes than to small,
dwellings. Designing architects and engineers must have sufficient know-
ledge of the construction possibilities by which an adequate fire resistance.
time can be attained in a given member.
In considering the behaviour of structural members in fire, two
basic cases must be distinguished. During a fire a construction is only
briefly, but very severely exposed. (For this unrepeated catastrophic case,
only a simple safety measure is required.) The state of the structure after
the fire, i.e. with a view to continued use, is only of secondary consid-
eration. For building compononents which have to withstand comparatively
high temperatures while performing their design functions, quite different
conditions prevail. This is the case, for example, in furnace construction,
and more recently in reactor design. Three different points of view must be
taken into account both in the selection of the building materials and in the
structural design.
In many respects there is no fundamental difference between the
behaviour in fire of precast and poured-in-place concrete construction. The
present paper reviews the behaviour of concrete and reinforced concrete in
fire and will discuss from individual examples the special properties of
concrete products, precast reinforced concrete members, and the posqibility
of improving the fire resistance time of individual members.
2. The Behaviour of Building Materials Exposed to Elevated Temperatures
2.1 General
With respect to the behaviour of building materials exposed to
elevated temperatures, we can separate two broad classes, combustible and
non-combustible materials. Wood and many plastics, for example, are
the "combustible" building materials. At higher temperatures, they begin to
burn and are destroyed. Not so well-known is the fact that the "non-
combustible" materials, including concrete and steel, suffer an alteration of
important mechanical properties as an effect of exposure to elevated temper-
-5-
atures. Regarding the behaviour of structures in fire, the following
IJrOperties of the building materials and their changes due to higher
temperatures are of importance:
1. Coefficient of thermal expansion;
2. strength, yield point, elongation at fracture and creep be-
haviour;
3. modulus of elasticity;
4. thermal conductivity and specific heat.
Moreover, changes in composition and structure of some building
materials can affect the behaviour of the construction ~ n a fire. For
example, many rocks crumble on exposure to extreme heat.
We shall now review the most important changes that occur in concrete
and building materials which are used in conjunction with concrete, due to
the effect of higher temperatures.
2.? Mortar and concrete
2.21 Thermal expansion
Mortar and concrete are made up of cement, aggregates and water.
The high temperature behaviour can differ considerably, depending on the mix,
the materials employed and especially the aggregates. While at room temper-
ature the mechanical properties of the solid concrete are determined
primarily by the properties of the cement paste*, the high temperature
behaviour, especially the coefficient of thermal expansion, is greatly in-
fluenced by the aggregates as well, since they make up the greater part of
the volume of the concrete. Therefore, let us first consider the behaviour
of the aggregates.
The effect of the aggregates on the behaviour of mortar and concrete
at higher temperatures depends primarily on their coefficient of thermal
expansion. The texture of a specimen of mortar and concrete can remain intact
in the presence of increasing temperatures only if the resulting changes of
shape of the separate components are mutually compatible, and the internal
stresses can be absorbed by the cement paste. Hence, aggregates with a low
coefficient of thermal expansion and a temperature-expansion curve that is as
smooth as possible, behave best. Figure 2 shows the thermal expansion of
* hydrated Port land cement paste.
-6-
various aggregates as a function of the temperature(S,6). It will be noticed
that for quartzitic rocks there is a characteristic discontinuity of the
thermal expansion between SOO and 600°C. This bend in the temperature-
expansion curve is due to the a ~ S quartz conversion, which under laboratory
conditions takes place at S73°C with a heat consumption of approximately
8 cal/g. To be sure, the practical effects of quartz conversion is not as
serious a matter in a fire as it may appear to be at first glance. Contrary
to experiments with small specimens, structures in the fire are only slowly
heated from the outside in to a temperature above SOOoC. Structural
relaxation, therefore, proceeds only to a limited depth. Of course, if
concrete constructions that have been exposed to fire are to be used again,
the affected layers must be r-emoved .
With the exception of diabase, non-quartzitic rocks generally behave
better than quartzitic material. Among natural rocks, basalt and a number
of kinds of limestone are found to have favourable coefficients of thermal
expansion, while artificial stones similarly endowed include blast furnace
slag fragments and crushed brick. Even for basalt, there is a c r i t i c a ~
temperature, at approximately 900°C. At this temperature the rock begins to
expand markedly, giving off gas.
When mortars or concrete specimens are heated it is not only the
aggregates that expand. The cement paste also changes in volume. Figure 3
shows the change of volume of cement paste from Portland cement according to
Endell(S), Nekrassow(6), and Philleo(7). The numerical data of these
investigators differ. It is clear, however, that first a thermal expansion
occurs, which is subsequently compensated for and exceeded by the shrinkage
due to release of water. A further reduction of volume takes place on
cooling after heating. Probably the different results are due to different
compositions of the cement paste (water-cement ratios) and different rates
of heating and different sizes of specimens.
Table I contrasts the various test conditions.
If the cement paste is heated not once, but several times, then,
according to ref.(6) , a reversible positive expansion occurs from the second
heating on (Figure 4).
Since the aggregates expand when exposed to heat, whereas the cement
paste begins to shrink again at higher temperatures, the thermal expansion
occurring in concrete is less than the aggregate itself. Under otherwise
similar conditions the coefficient of thermal expansion decreases with
increasing cement content(7). In Figure S the thermal expansions of various
-7-
mortars are plotted as functions of temperature after Endell(S). All stones
of the same kind do not necessarily behave the same. Considerable variations
are possible among quartzitic material, and especially among limestones.
2.22 Strength
When specimens made from mortar and concrete are exposed to higher
temperatures their strength values are altered. Publications(6,8-1S) on
strength tests of mortar and concrete in a heated state are partly contra-
dictory. This is due on the one hand to varying test conditions (sDecimen)
size, mix, type of aggregate, age of the specimens, storage conditions,
heating times, time of testing after heating, water-cement ratio, construction
of test furnaces, etc), and secondly to differences in the initial material.
The investigators fall into two groups. One group already found a decrease
in compressive strength for the comparatively slightly heated stage of about
200°C, which continued at a more or less steady rate at higher temperatures.
Other tests, however, showed an increase of compressive strength up to about
300°C and a drop only on further heating. The initial increase of strength
on heating of certain concretes can be explained by the fact that heating
brings about a "steam hardening", leading to additional hydration of the
cement. With many types of aggregates, moreover, a reaction between cement
and aggregate is by no means impossible.
The vertically hatched area in Figure 6 shows the scatter range with-
in which most of the results of concrete compressive strength tests at higher
temperatures fall. The data of Malhotra(ll), according to which the water-
cement ratio has no appreciable effect on the relative change of thermal
compressive strength, are interesting. Of greater influence, however, is the
ratio of cement to aggregates, which presumably can be attributed to the
above-mentioned severe internal stresses arising between cement pastes and
aggregates.
British investigations(16) showed that after exposure to temperatures
up to about 300°C for several weeks no further decrease of strength could be
expected (Figure 7).
According to French investigations(17), concrete tensile strengths
drop more rapidly than the compressive strength values. The scatter range
for the tensile strength is horizontally hatched in Figure 6. For a special
fireproof concrete, the compressive strength may be greater after very
severe heating than those indicated in Figure 6(6,12,lS). The increase of
strength in the temperature above lOOOoC, represented in Figure 6 by the
-8-
broken-line curve, is due to the fact that in properly composed concrete the
lost bond of hydration is replaced by a ceramic bond.
2.23 Mortulus of elasticity
When concrete is exposed to elevated temperatures, a decrease in the
modulus of elasticity occurs. This strongly affects the deformations in a
structure. Also affected is the interplay of external and internal forces in
statically indeterminate structures, in prestressed concrete constructions
and even in standard, reinforced concrete cross-sections.
Some existing results of modulus of elasticity measurements at
higher temperatures differ sharply from each other. Figure 8 shows
Woolson
1s(18)
results on diabase and limestone concrete, later confirmed
by Busch(9) on sandstone concrete (quartzitic). Woolson investigated
concretes with a cube strength of about 160 kg/cm
2
• Busch's investigation
dealt with cube strengths of 280 kg/cm
2
• The compressive stresses were 30
to 50 kg/cm
2
• The very strong decrease of the modulus of elasticity even at
relatively low temperatures up to 400
0
c is striking. According to Woolson
and Busch, the modulus of elasticity at 400
0
c is only about 20% its value at
room temperature.
In more recent investigations(7,12,19) a slower rate of decrease of
the modulus of elasticity was found l)y Cruz and Philleo. The results of these
investigations are also plotted in Figure 8. However, in these tests, also,
the modulus of elasticity at 300
oC,
again fell to 40 to 50% of its value at
room temperature, depending on the water-cement ratio.
The difference between the results of Woolson and Busch on the one
hand and those of Cruz on the other may be sought in the fact that according
to Busch's findings the modulus of elasticity of a concrete that has been
exposed to fire increases with increasing load. This means that the ratio
of E Id/E is more favourable when the tangent moduli are compared, not,
co warm
as usual, the secant moduli. For practical application, therefore, values
may be chosen which lie between those of Busch and Cruz, depending on the
nature of the stress.
2.24 Heat conduction and specific heat
The internal heating of a structural member depends largely on the
heat conductivity and the specific heat of the building materials employed.
These are not constant values, but depend on various factors. If we con-
Range I
Range II
Range III
-9-
sider the heat conduction of concrete, we must distinguish between three
basic temperature ranges:
be10w oOe
between 0° and 1000e
above 1000e.
In conjunction with the behaviour of concrete structures in the
presence of fire, range I is of interest only in a very few exceptional
cases. Normally, the initial temperature of a construction at the start of
a fire will be above oOe. For the sake of completeness, however, it must be
mentioned that a considerable delay in the internal heating of the structural
members is due to the latent heat of melting of ice.
In range II the heat conduction (kcal/mhOe) of the concrete is
influenced primarily by its bulk density and its moisture content. Figure 9
shows these effects at room temperature(20). The specific heat of regular
air-dry heavy concrete at room temperature can be taken as 0.21 - 0.25 kcal/
kgOe.
In Figure 10, with the example of an aerated concrete having a
bulk density R = 520 kg/m
3
, it is shown that the thermal conductivity at
constant bulk density depends not only on the moisture content, but also on
the temperature(21). At higher temperatures the heat conduction characteris-
tics increase. This is due to the fact that at higher temperatures the
amount of heat transferred by radiation in the pores increases greatly. It
may also be inferred from Figure 10 that the maximum heat conductivity value
of aerated concrete is attained not for total saturation with water, but
with partial filling of the pores.
In temperature range III, concretes are completely dry. The
coefficient of thermal conductivity, therefore, is less than in range II. In
Figure 11 the thermal conductivities of a few concretes at higher tempera-
tures are represented as a function of the temperature(22,23). Here again
there is a definite rise in the thermal conductivity with temperature.
The practical importance of the increase of thermal conductivity of
concrete with the temperature is limited. The heating of a thick plate
subject on one side to temperatures which vary in time is governed by
Fourier's equation of heat motion. This is written
6T 2 6
2T
IT = a 6s
2
(thermal diffusivity)
-10-
where
T temperature (OC)
t time (h)
s = depth in the wall (m)
2
a
2
coefficient of thermal diffusion ( ~ ). (thermal diffusivity)
For heat transfer, therefore, the decisive factor is not the
thermal conductivity, but the thermal diffusivity a
2
, which is defined as
follows:
where
2
a
2
= coefficient of heat diffusion ( ~ )
thermal conductivity (: kcal )
c = specific heat (kCal )m· h- °c
kg . dC
Y bulk density (kg/m
3
)
Since the specific heat c of the concrete also increases within
increasing temperature(9,12), the thermal diffusivity which depends on the
quotient }, varies but little.
Much more decisive is the delay in internal heating in a fire owing
to the evaporation of the water in the concrete. The latent heat of evapo-
ration of water, of course, is over 500 kcal/kg. This is a very large amount
of heat. It follows that the behaviour of a concrete construction in a fire,
and especially its internal heating, is decisively influenced by the water
content.
A parallel phenomenon occurs in limestone concretes. Here quantities
of heat are also consumed in the conversion of CaC0
3
into CaO and CO at a
2
rate of nearly 400 kcal/kg CaC0
3
at a surface exposed to the fire, thus
delaying the heating of the structure. The spontaneous liberation of CO
2
by the limestone sets in at approximately 900°C, but at low CO
2
partial
pressures it may start as low as 600°C.
Figure 12 shows the loss of weight of a concrete containing some
calcareous aggregate according to ref.(7). The vigorous liberation of water
between 100 and 300°C is clearly evident. Above 400°C a slight loss of
weight then occurs due to liberation of water from the calcium hydroxide of
the cement paste. At about 600°C the conversion of the limestone begins.
This may be summed up as follows:
The thermal conductivity of concrete is determined primarily by the
bulk density and the moisture content. The greater the bulk density and
moisture content, the greater the conductivity. The complete heating of
-11-
structural members is affected decisively by the evaporation of the free and
bound water.
2.3 Steel
The fire resistance of all reinforced concrete constructions,
including steel concrete and prestressed concrete, is decisively affected by
the behaviour of the reinforcement. The change in the mechanical properties
of steel on exposure to high temperatures depends on the kind of steel and
especially on the carbon content. In the following paragraphs a number of
typical examples are cited inasmuch as these are required for an understanding
of the behaviour of reinforced concrete constructions.
2.31
Thermal expansion
In Figure 13 the linear thermal expansion of a steel with 0.4% carbon
is plotted after ref.(24). It will be recognized that the thermal expansion
is continuous up to temperatures of about 700°C. The coefficient of thermal
expansion increases slightly. Compared with the thermal expansion of the
concretes (Figure 5) the expansion of steel is somewhat less than sand-and-
gravel concrete, but somewhat greater than limestone concrete. Generally
speaking, however, the thermal expansion of the steel and the concrete is the
same even over considerable temperature ranges.
2.32 Yield point at elevated temperatures
Figures 14 and 15 show the yield point (0.2 proof stress) for a
number of steels as a function of the temperature. The hot tensile strength
curves have been left out, because by and large they are closely related to
the yield point curves. Only in the temperature range around 200°C is there
at first a slight rise in the tensile strength.
Figures 14 and 15 differ in that in the former the tests were carried
out in the hot state, while in Figure 15 they took place after cooling. The
values of Figure 14 are important for the behaviour of the material during
a fire, those of Figure 15 for appraising the bearing capacity of a structure
after a fire. Rolled steels in general recover their original strength on
cooling, whereas cold-forged steels lose some of their strength.
It is clear from Figure 14 that the yield point decreases at first
0.11
-12-
slightly with increasing temperature, but then more rapidly. At 500 to
o
600 C the design stresses calculated in the construction are attained.
Contrary to a widespread misconception, the yield point of high
quality prestressed rods is not below that of ordinary reinforcement rods.
It is only in the temperature range of around 600°C that all the values more
or less coincide.
2.33 Rupture strain
Figure 16 shows the change in rupture strain of a mild steel under
the influence of elevated temperatures, after ref. (27). Up to 200°C the
rupture strain at first decreases rapidly, but then in the region of higher
temperatures it increases just as rapidly. At 600°C the failure strain is
about twice as great as at 20°C.
2.34 Modulus of elasticity
Figure 17 shows the change in the modulus of elasticity of a non-
siliceous S. M. mild steel as a function of the temperature. The behaviour
of other steels is similar. It is clear from Figure 17 that the modulus
of the steel drops to about half its value at room temperature between 500
and 600°C. The decrease, therefore, is not as great as in concrete.
2.35 Thermal Conductivity and specific heat
The thermal conductivity of the steels normally used in construction
decreases with increasing temperature. Figure 8 shows the heat transfer
coefficient of pure iron and of structural steel with 0.8% carbon content.
The values for most structural steels lie within the hatched area. It must
be realized that at comparatively low temperatures, i.e. at the beginning of
a fire, the heat transfer coefficients of steel are 30 to 50 times as great as
that of concrete. This fact affects the behaviour of very heavily reinforced
structural members in the fire.
The specific heat of iron and steel increases with increasing
temperature. At room temperature the specific heat is approximately c
kcal
kg .uc
and at 800°C it is 0.2 kcal
kg.oC
2.4 Other materials
In the erection of concrete structures, especially where pre-
-13-
fabricated concrete parts are used, a number of other materials besides
concrete and steel may be employed in the finished structure, e.g. as
bonding veneers, plasters or for sealing purposes. These materials
include, for example, gypsum products, wood or wooden wares, and plastics.
In what follows we shall briefly outline the most important prop0rties of
such materials with respect to fire.
Gypsum, as is known, hardens by hydration of the semihydrate or of
the anhydrate to the dihydrate form. When gypsum is exposed to higher
temperatures, some water is given off between 100 and 160°C and complete
dehydration to anhydrate occurs at about 200°C. For this, considerable
quantities of heat are consumed. Hence the heating of all constructions
made of gypsum is greatly delayed at approximately 100°C. Use can be made
of this fact at times, when rapid heating of a building element must be
avoided.
However, after dehydration the gypsum loses much of its strength.
Moreover, with gypsum plasters under very gas-tight floors the steam
pressure may become so great that the entire plaster breaks away from the
ceiling.
Wood occurs in concrete structures, e.g. in the form of small
structural parts such as dowels, or the like. Wooden laths are also embedded
in the concrete in the manufacture of certain concrete products with a view
to securing projecting sections. Wooden parts surrounded by concrete on
several sides do not ignite as easily as completely exposed ones, owing to
their poor contact with the atmospheric oxygen. In addition, larger cross-
sections always have a higher resistance to fire than delicate, thin
structural parts. Caution is always required whenever parts of the whole
construction which are essential for fire protection are secured to wooden
parts. A critical temperature for the ignition of wood can only be stated
approximately, since it varies greatly with the kind of wood, its moisture
content, the atmospheric oxygen available, the surface texture and other
factors. Under unfavourable conditions, however, ignition may be expected
at 250 to 300°C. Some times, even lower ignition temperatures have been
noted.
Ceiling fillers or slabs of mineralized wood wool have been found
excellent in conjunction with concrete. The resistance of thin steel re-
inforced concrete ribbed ceilings to fire can be greatly improved by the use
of these parts, since even in a fully raging fire they ash slowly, and thus
constitute good
In the field of plastics, development is in full flood. No general
-14-
critical temperature for all plastics can be stated. Some plastics alter
their properties even at lower temperatures than wood. Especially in the
field of prefabrication, where plastics are popular as binding agents, ad-
hesives or joint seals, therefore, fireproof requirements must be taken into
account in the planning stage, because subsequent changes are frequently
very difficult to carry out.
3. The Behaviour of Concrete and Steel-Reinforced Concrete Members in Fires
3.1 General
A knowledge of the change of properties of building materials will
suggest important ideas concerning the behaviour of structures in fire. The
fire test on the whole structural part, however, is decisive in appraising
fire resistive quality. In many countries there are standards for conducting
fire tests. These regulations are alike in principle, although, they differ
from each other in detail.
Basically, it is laid down that the structural part to be investiga-
ted must be exposed to damaging fire(3
0),
i.e. a fully developed conflagration.
In order to get comparable test results it is customary to represent the
development of an idealized fire by temperature-time curves. Figure 19
shows the temperature-time curves used in various countries. It will be
noted that all the curves are very similar to each other, so that the results
on constructions investigated by the standards of various countries on the
whole are comparable with respect to temperature load. In order to get the
best possible agreement, a uniform curve has been drawn by ISO, which fits
the previous curves well. However, even under earlier conditions, differences
in the results from fire tests in different countries have been due less to
different temperature loads than to differences in test material and other
test conditions.
In attempting to analyse the curves represented in Figure 19 it must
first be realized that in the first half hour, at the start of the fire, the
temperatures in the fire compartment rise very rapidly, and then increase more
slowly. The rapid changes of temperature at the beginning of a fire in
general result in large temperature differences in the structural member. At
an early stage, therefore, considerable deformations or stresses occur. As
the fire continues, there is a continuously progressive, but more even
heating of the construction, which, unless a temperature equilibrium sets in
before hand, can result in failure. The uniform temperature-time curves
correspond approximately to the course followed by a fire of medium severity.
-15-
This is important, because for the most part the temperatures lie below
the 1200
0C
limit. At this temperature, however, decomposition or de-
struction of a number of mineral building materials occurs. Several
kinds of natural stones, for example, decompose. Many light-weight
concretes, which show excellent behaviour in fire up to about 1200
0C
owing
to their good thermal insulation properties, are destroyed when the
temperatures go higher than this. Of course, such conditions would occur
only in rare and exceptional cases.
In the course of a fire the structural parts ought to perform their
functions. Essentially, these are as follows:
a) Bearing parts, such as beams, columns, ceilings, etc. must retain their
bearing strength and stability.
b) Space separating parts must continue to be effective in separating space,
i.e. they should not afford a passage for the fire.
Basically, two cases must be considered in the testing of structural
members for their resistance to fire. On the one hand, the structures may
be tested to failure, i.e. until they no longer satisfy the above functions
a or b, as the case may be, or a time test may be made, i.e. it is
determined whether the functions will continue to be performed during a
predetermined time.
In order to be able to estimate whether and how long the bearing
capacity of a structure will remain intact in a fire, the resistance of
bearing structural parts to fire is tested under load. There is little
difference in the regulations on this point from country to country. In
general, however, the simple calculated permissible load is applied in the
fire test. Since fire is a catastrophic occurrence, a safety factor of one
is regarded as adequate as far as bearing capacity is concerned.
The stability is threatened by severe deformations. For example,
in the case of walls exposed to the fire, on one side, the resulting de-
flection may produce such great eccentricity that the wall will collapse.
Space separating structural members must not allow the fire to pass.
That is to say, no hot gases should seep through, and the temperatures on
the side away from the fire must not rise above a certain value. The
critical rise in temperature in most countries today is put at an average
of about 140°C above the initial temperature, and at individual points up
to 180°C is allowed. The reason for this requirement is to prevent easily
ignitable material, which may by chance be stored on the unexposed side of
the space separating members, from igniting, for otherwise the fire will
-16-
jump the barrier.
In a number of countries some relaxation of these rules is allowed
for certain members, e.g. doors, parapets, or walls of glass blocks.
Special demands are made of chimneys and roofing materials.
The duration of the fire test depends on the type of structure in
which the test member is to be used. Requirements differ greatly in the
various countries. In Germany for buildings threatened with fire, "fire-
resistant", and in exceptional cases "highly fire-resistant" construction
is reqIJired. "Fire-resistant" members, according to DIN 4102 must continue
to perform their functions in the fire tests for 1 1/2 hours, and "highly
fire-resistant" must do so for 3 hours. In other countries, for certain
buildings such as warehouses, which are particularly exposed to fire
hazards, still longer resistance times are prescribed.
Increasing the demands for duration of resistance against fire
beyond a certain time does not always require substantial changes in the
construction. This is because for many structural members there is a
certain time at Which, during the standardized fire test, heat equilibrium
is more or less attained. In this state the heat losses on the side away
from the fire are approximately equal to the input of heat on the side
towards the fire. If the structure has not collapsed by this time its
resistance of fire can be greatly increased by reaching this threshold.
To carry out the fire test, the structural member to be tested is
installed in a fire test chamber, and, if it is a bearing is subjected
to a load. Figure 20 shows a fire test chamber with a loading apparatus
for walls. The wall is prepared for testing. Figure 21 shows, semi-
schematically, a fire test chamber for ceiling constructions. Today the
testing institutions of many countries have very modern fire test chambers
with partially or fully automatic control.
The test member is installed in the fire test chamber in a manner
that simulates as closely as possible the incorporation of the member in
practice.
3.2 Walls of light weight and heavy concrete
3.21 Heating
Concrete walls can be manufactured either monolithically from heavy
or weight concrete, or may consist of slabs of considerable size or of
blocks. Basically, we must between bearing and non-bearing
-17-
walls. The former have both bearing and space separating functions, where-
as the latter operate only as space dividers. In general, thick walls of
the kind required for outside walls in tall building constructions, due to
thermal insulation considerations are not endangered by fire. In the case of
thin walls, however, there is a possibility that the permissible rise in
temperature on the side away from the fire may be exceeded.
Figure 22 shows the necessary thickness for massive walls of heavy
and light weight concrete as a function of the test time, so that no in-
crease of temperature ~ r e a t e r than 140
0
c above the initial temperature will
occur on the unexposed side. In the case of walls with cavities, the
deciding factor is the thickness of solid material. The fire resistance
time F increases exponentially with the wall thickness d, approximately
according to the function F = d
l.
7
( 3
1).
The critical increase of temperature on the side away from the fire
up to test times of 3 hours is only important, generally speaking, in the
case of heavy concretes and wall thicknesses of d < 15 cm. A few kinds of
hollow blocks made of heavy concrete constitute an exception to this. As
in the case of hollow blocks of other materials, in unfavourable cases it
can happen that the shells on the side of the fire will burst after a
certain time and that cracks will form through the entire block as the test
continues. The permissible rise in temperature for light-weight concretes
is seldom exceeded, even in the case of thin walls.
For masonry brick walls or slab walls, careful design and execution
of the joints is extremely important. Grouting or smoothing of the joints
may help, so as to ensure that no hot gases can penetrate through any gaps
in the joints. Especially where thin plates are employed, the joints have
to be profiled as far as possible, so that adjoining plates fit together in
tongue and groove fashion.
The thermal insulation of the structural members during a fire is
greatly affected by the moisture content. First of all, considerable
quantities of heat are needed for the evaporation of the excess water,
which quantities, of course are liberated again during condensation. Thus
structural members with high moisture content rise rapidly to lOOoe, but
only slowly go above lOOoe. In the case of concrete separating members,
such as walls and ceilings, lags i ~ the temperature rise are observed on
the side away from the fire generally for temperatures between 80
0
and
1000e. Moisture often penetrates in the form of water of condensation
during a fire at joints and hairline cracks. The temperature of these moist
places is somewhat higher than that of their dry surroundings. In the case
-18-
of a heavy concrete wall with a bulk density of about 2400 kg/m
3
, a
difference of 1% by weight in moisture content results in temperature
differences of about 25°C on the unexposed surface, after all moisture
has evaporated.
In Figure 23 the temperature curve in heavy concrete walls is
represented as a function of the wall thickness, and in Figure 24 the same
curve is shown for a 15 em-thick wall of aerated concrete. The two
diagrams are indicative only, since the scatters and temperature measurements
inside the walls are very great owing to differences in moisture content,
aggregates and bulk densities. The delay in heating at 100°C is definite
in the case of the aerated concrete wall, as is also a change in thermal
conductivity above and below 100°C.
Numerous compilations about fire tests on walls are available in
the literature. Table II contains a selection of a few typical walls.
Figure 25 shows a wall of slag concrete blocks after a 3-hour fire test.
3.22 Deformations and stability
When exposed to fire on one side, walls tend to bulge more or less
extensively towards the fire side. After the fire, the bulging generally
recedes, and the bulge may even appear on the other side. The magnitude of
the bulging depends on the material, the wall thickness, and above all on the
height of the wall. The mean bulging of unfixed walls can be estimated
approximately by the following equation
f
- C h
2
(Tl - T2)
- 8d
where
C - material constant
f - deflection
h - height of wall
T - temperature on the side towards the fire
1
T - temperature on the side away from the fire
2
From this equation the important fact is realized that under other-
wise equal conditions the bulging increases with the square of the height
of the wall. In Figure 26 the conditions are represented for two walls
which are similar except for their height. One wall is twice as high as the
other. The bulging on one side, however, is four times as great. This must
be pointed out, because fire tests on walls are usually made only on spec i-
-19-
mens 2 m to 3 m high, i.e. the usual height of a storey. In the case of
taller walls, which indeed are rare, but which do exist the
effect of the greater bulging in fire should be taken into account, because
stability may no longer be guaranteed owing to large eccentricity.
Figure 27 shows the bulges in two walls during fire tests.
Besides lateral bending, gaps may occur in the joints at the margins
owing to bulging of the walls. Therefore, good 'connection between the wall
and the adjacent structural members must be assured in the design.
3.3 Floor structures
3.31 Massive, steel reinforced concrete slabs
Whereas walls are predominantly compressively stressed, steel-
reinforced concrete floors as a rule are sUbject to bending stress.
Generally speaking, in floors of tall buildings there is tension on the
underside of the slab and compression on the upper side. In a fire, a
ceiling above the burning area is rapidly heated on the underside. Owing
to the relatively poor heat conduction of concrete the steel reinforcement
situated close to the underside in the tensile zone is heated more rapidly
to the critical temperatures of steel than is the concrete on the upper
side of the floor structure heated to its critical temperature. Therefore
the heating of the steel reinforcement plays a decisive role in the
resistance of steel reinforced concrete floors to fire.
The time during which a steel reinforced slab resists the fire is
determined substantially by the following factors:
a) existing steel stress
b) steel covering
c) nature and thickness of lining underneath
d) quality of concrete
e) thickness of the floor
f) moisture content of the concrete
g) kind of aggregate
h) positioning of the slab (restraint)
Figures 14 and 29 show clearly that reducing the steel stress, or
in other words over-reinforcing, increases the length of time of the
resistance.
By increasing the covering of the steel, the heating rate of the
steel reinforcement is reduced and thus the resistance of the steel re-
-20-
inforced concrete slab to fire is increased. For a slight increase in the
steel covering, for example from I to 2 cm, the increase in the time of
fire resistance, to be sure, is not very great. The reason is evident
from Figure 28. It can be recognized that in the side of a wall or ceiling
nearest the fire there is a very steep temperature gradient, which becomes
flatter towards the inside. That is to say, the temperature difference
between the underside of the floor exposed to the fire and the depth of I cm
is greater than between I cm and 2 cm depth from the surface. A substantial
increase in the time before fracture is obtained only with still greater
steel covprings. Plasters in earlier tests showed greater protective
effect than corresponding increases in steel covering. An essential
condition, however, is that the plaster adheres well. The explanation may
be sought in the fact that plaster has a lower bulk density than concrete
and thus shows greater thermal insulating properties. As a consequence the
temperature gradient in the vicinity of the concrete ceiling surface on the
side of the fire is steeper than that in Figure 28. Moreover, heat
conduction is hindered by the transition between different layers. On the
debit side it may be said that in practice plasters are frequently applied in
thinner coats than contemplated in the design and that green plasters with
high moisture content are quickly stripped off in a fire. This is especially
true if the underside of the concrete is smooth. Since the test described
in ref. (25) was made, concrete technology has advanced. Today, in general
more compact concretes are produced. It is therefore easy to understand
that in more recent fire tests the favourable effects of plasters are not
always evident, since the plasters have been detached from the concrete at an
early stage in the fire. Priority should be given to this problem in the
future, since in prefabricated structures traditional plasters may be used
less than before, for other reasons as well.
For a number of years special coatings, e.g. of asbestos,
vermiculite and perlite bases, have been used to increase fire protection.
With these coatings, even comparatively thin ones, the heat insulation can
be so greatly improved that the fire resistance time is increased to many
times that of the uncoated floors. F i ~ u r e 29 gives two examples showing
the effect of coatings on the heating rate of the steel reinforcement rods
and hence on the fire resistance time of reinforced concrete floors. The
curves may be designated as follows:
Curve I - uncoated steel reinforced concrete floors (covering depth
I cm) after ref. (25);
-21-
Curve 2 - the same construction, but with a lime-cement plaster
1.5 cm thick;
Curve 3 - the same construction, but with a cement-lime-vermiculite
c o a t i n ~ 1.5 cm thick.
The effect of concrete quality on the fire resistance time has not
yet been fully clarified. In earlier tests it was established that for
poor and medium quality concrete up to approximately B 300 the ability of a
steel reinforced concrete floor to resist fire increased somewhat with
increasing concrete grades. These results, in more recent tests, which
were not always confirmed are somewhat surprising, since the bulk density
and the thermal conductivity of the concrete in general increases with
increasing grade. Perhaps the explanation is that with the increase in
concrete grade the modulus of elasticity also increases. Given equal
thickness and reinforcement the floor, however, a higher modulus of
elasticity of the concrete means a lower steel stress. In going from a B 160
to a B 300 the modulus of elasticity increases by approximately 50%. This
displaces the neutral axis of the cross-section upwards, and, for a concrete
floor 12 cm thick with an average reinforcement component, causes a drop in
steel stressing of about 6 to 10% (Figure 30). With an unplastered ceiling
this corresponds to an extension of the fire resistance time by about 10
minutes. However, these findings for comparatively low grades of concrete
must not be applied to higher grades, because here the modulus of elasticity
does not increase greatly, and secondly very compact concretes do not be-
have at all favourably in fires, and under the most unfavourable conditions
tend to spall.
The considerable effect of the thickness of the floor on the fire-
resistance time is represented in Figure 31. This relates to concrete slabs
reinforced with comparable steel rods. Aside from the fact that
comparatively large cross-sections behave more favourably in fire owing to
their greater heat capacity, it may also be assumed here that despite
mathematically equal calculated steel stress, the actual steel stress in
a thicker floor is less. Another favourable result is that the changes of
form of a thick floor are comparatively less than those of the thinner one.
Moreover, it should be borne in mind that under otherwise equal conditions,
the moisture content of a thicker member is generally greater and hence
it takes longer to heat through. As will be shown later, the fire resistance
times in other structural members, also, are increased by increasing the
cross-sections. The influence of floor thickness on fire-resistance time is
31 minutes
41 minutes
65 minutes
2 hours
-22-
of special importance in practice, because in Germany almost all relatively
extensive series of fire test have been carried out on steel reinforced
concrete floors of thicknesses less than 10 cm. The values obtained formed
the basis for the pertinent regulations. However, for the thicker floors
now being used predominantly in practice, these values are much too unfavou-
rable. This gain may be particularly important especially in prefabricated
construction, where for a great variety of reasons plastering on the under-
side of steel reinforced concrete floors is not possible. A fire-resistance
time of 90 min ("fire resistant") is readily attainable with floor thick-
nesses of about 15 cm, increasing the steel covering from 1 to 2 - 3 cm.
The static system also influences the bearing capacity of a floor in
a fire. Continuous slabs are more favourable than single bay slabs. In
the case of continuous slabs, before failure of the strongly heated horizon-
tal reinforcement, at first the still relatively cool upper column reinforce-
ments act with increased stresses as cantilevering reinforcement. As a
consequence there is a relieving of the stresses in the bay reinforcements.
Collapse does not occur until the yield point is exceeded in lower and
upper reinforcements. From a fireproofing point of view, therefore, a
continuous upper reinforcement throughout the bay is especially favourable.
The behaviour of single-bay slabs fixed at all sides has not yet
been fully clarified(3
4
, 67) .
Table III gives a summary of the increases of fire-resistance time
as a result of the above-mentioned measures.
Besides these possibilities, the resistance of massive steel-
reinforced concrete slabs can also be affected by the choice of aggregates.
To be sure, the available data do not provide a uniform picture, since
obviously other factors, for example, moisture content at the time of the
test, the restraint of the slabs in the experiment, etc., have a very
strong influence on the test results. According to Davey and AShton(3
4),
concrete floors of similar nature but made with different aggregates had
the following fire resistance times:
Flint
Dolerith
Basalt
Crushed brick
-23--
3.32 Prefabricated floors
3.321 Fire-resistance time
In recent years prefabricated floors have been very widely used.
Among these a few basic types may be singled out from the very large
number of systems found on the market.
1. Beam floors with intermediate structural members.
2. Ribbed floors with intermediate bearing members or compressed
slabs of poured-in-place concrete.
3. Floors consisting of beams or planks laid side by side.
4. Prefabric2ted slab floors.
The behaviour of pr-e f abr-Lc a t e d floors in f l i-e Lr: principle is
similar to that of solid slabs. Since the steel reinforcements are frequent-
ly better protected than irl solid slabs, in many cases a longer fire-resist-
ance time may be expected than solid slabs of equal thickness.
Accordingly the measures indicated under point no. 3.31 for solid
slabs for improving the fire-resistance time hold also for prefabricated
floors. Investigations have shown that almost all non-prestressed reinforced
concrete floors which were given a coating of lime-cement 1 1/2 cm thick on
the underside survive the required test time of 1 1/2 hours in the fire
test according to German regulations.
In Table IV, examples of typical prefabricated floors are given with
the corresponding test results. For comparison, the results of a few solid
floor slabs are also listed.
3.322 Deformations
Besides bearing capacity, changes of shape in the course of a fire
are also important in the behaviour of prefabricated floors. Directly
after the start of fire exposure, floors exposed to the fire begin to change
their shape. As a rule they bend toward the fire, i.e. they begin to sag
downward. Figure 32 shows the deflections of some of the prefabricated
floors listed in Table IV. The value of the deflection depends not only
on the span and thickness of the floor, but also on the construction and
the material. For examole, hollow beams of heavy concrete or solid light-
weight concrete slabs bend somewhat more than solid heavy concrete slabs
under similar conditions.
In the case of prefabricated floors of beams or planks laid side by
-24-
side, we warlted to know how they would behave in the presence of fire under
partial If the deformation of loaded parts of a floor compared
with that of the unloaded parts, there is a danger of shearing at the
senaration joints and the fire breaking through. A number of floors were
therefore in fires, only half of which were loaded on one side
with the calculated permissible loads(43). All the floors (floors without
transverse reinforcement and with relatively large deflection were deliber-
ately chosen for the test) resisted the fire under these conditions and the
fire did not break through. In addition, small floor units of aerated
concrete were tested in unloaded and loaded conditions. As is evident from
Figure 33, the results showed that the deflections were almost the same at
the beginning of the fire test. This means that the initial deflection
depended entirely on the temperature difference between upper and lower
sides. Only after exposure to the fire for over 20 minutes did the deflec-
tion of the loaded floor begin to increase more strongly than that of the
unloaded. At this time a gradual reduction of the modulus of elasticity
began. Even at this time, however, the deflection component due to temper-
ature differences remained relatively large compared with that due to the
loading.
In certain exceptional cases, floors are deflected at the start of
a fire not towards the fire, but away from it. This happens, for example,
if the ribs of a ribbed floor are heavily insulated in order to protect the
steel from heating, and the upper compressed slab heats up more rapidly than
the ribs. On the other hand, a permanent upward deflection after the termi-
nation of a fire is comparatively frequent. This occurs, for example, with
some aerated concrete floors. As a result of fire exposure the concrete
after cooling undergoes a slight, permanent increase of volume, while the
steel reinforcements revert to their original length. The result is a "pre-
stressing" effect which causes an upward curvature.
3.323 Downward propagation of fire
Generally speaking, fire tests are carried out under the obvious
assumption that fire spreads upward from below, i.e. floors above the fire
are mainly endangered. Sometimes, however, we are interested in the
behaviour of floors where the fire is propagated downward. For this purpose,
tests were made with a partial prefabricated floor of steel reinforced
hollow planks laid side by side(44). The same floor had already been tested
in the usual way (cf. Table IV, No. The tests showed that the floor
-25-
remained intact under fire exposure from above during the test time of
1 1/2 hours. The permanent deflection of the floor after the fire was
greater than in the regular tests. Figure 34 shows the upper side of a
concrete plank of the test floor after the fire.
3.324 Prefabricated slab floors
In the prefabricated, large-area steel reinforced concrete floors,
which have very recently come into use in apartment houses and in industrial
construction, it is often not possible to apply a plaster to the underside
of the floor during the prefabrication process. Subsequent plastering may
nullify the economy of this method of construction. Under these circum-
stances the fire-resistance time can be increased by the following
measures:
1. By decreasing the reinforcement stress
2. By increasing the reinforcement covering
3. By the use of suitable aggregates
II. By incorporation of an ancillary layer of lightweight concrete
with low thermal conductivity
5. By incorporation of slabs of heat-insulating materials either
in the concreting stage, or subsequently, over the finished and
installed ceiling.
Simply increasing the reinforcement covering is of limited value.
An unplastered high-grade concrete floor sometimes fails because one or more
pieces of the concrete surface spall off and at these points the steel re-
inforcements rapidly become heated. Structural measures should therefore
be taken, e.g. by incorporation of a light wire mesh, to ensure that spalled
portions cannot falloff. At the very high t e ~ p e r a t u r e s occurring in a fire,
heat transfer takes place primarily by radiation, and therefore particles of
concrete that have loosened but do not fall still provide adequate protection
for the steel reinforcements.
Ancillary layers are made mainly of lightweight concrete or from
slabs of other heat-insulating materials. The required fire protection has
also been attained in prefabricated concrete slabs simply with the aid of
gypsum boards or cement-bound wood-wool lightweight boards which are merely
concreted over (cf. Table IV, no. 2).
-26-
3.4 Prefabricated steel reinforced concrete steps
Steps constitute a special case of the ;'floor structure". Opinions
differ as to whether the behaviour of steps in a fire must meet the same
requirements as floors, i.e. whether they must remain effective as space
separators and at the same time retain their bearing capacity, or whether
only the latter is important. At the present time German regulations
require both specifications to be met. However, an amendment may be ex-
pected.
Prefabricated steps are generally produced in the following struc-
tural forms:
1. Large plates with steps mounted on them
2. Beams placed side by side
3. Side walls with steps placed between them
4. Beams with steps mounted on them
Basically the same considerations apply from a fire standpoint for
steps as for floors. Figure 35 shows examples of two designs which can be
regarded as "fire-resistant" according to the existing German regulations,
i.e. they have a fire resistance-time of more than 1 1/2 hours(45,46).
3.5 Steel reinforced concrete beams
Test results on the behaviour of steel reinforced concrete beams in
fire are available from various countries(25,34,47). In some instances
T-beams, and in others beams of rectangular cross-section, have been in-
vestigated. From these tests it may be assumed that the behaviour of steel
reinforced concrete beams is determined by substantially the same quantities
as thgt of slabs. However, no direct comparison between steel reinforced
slabs and beams is possible, since in the case of beams the ratio of the
surface attached by the fire to the cross-section is less favourable than it
is for slabs. On the other hand, in the case of beams the greater structural
height is a positive factor.
The effect of the steel covering has been investigated in consi-
derable detail. In the case of beams, it was again found that the fire-
resistance time did indeed increase with increasing cover, but not pro-
portionally to the thickness of the covering. Figure 36 shows the effect of
steel reinforcement covering on the fire-resistance time. In all the tests,
which owing to the different sizes of beams and different bearing arrange-
ments are not directly comparable, it was clearly evident that the increase
-27-
of fire-resistance time is less than the increase of steel reinforcement
cover.
The age of the member influence the fire-resistance time of T-beams
more definitely than plates(25). In Figure 37 the effect of age is repre-
sented. Presumably the increase of fire-resistance time with increasing
age can be attributed to a simultaneous increase of the quality of the
concrete and the modulus of elasticity.
As in the case of slabs, an increase of fire-resistance time can
also be obtained by a suitable choice of aggregates(3
4
) . Three beams were
produced which differ only with respect to the kind of aggregates employed.
Figure 38 shows the of the beams; the test results are contain-
ed in Table V.
In considering the earlier test results on beams it should be
mentioned that the beams were made of relatively low-strength concretes.
Nowadays higher grade concretes are generally employed, especially in the
manufacture of prefabricated steel reinforced beams. With structural
members of high quality concretes there is a danger of the explosive spalling
of pieces of concrete from the surface. If parts of the beam reinforce-
ments are thus exposed, rapid failure occurs. It is assumed that this
spalling is caused by water vapour over pressure due to evaporation of the
water present in the concrete. This view is supported by the fact that such
spalling occurs predominantly in green and very high quality, i.e. very
compact concretes, shortly after the start of exposure to the fire.
Probably other influebces also contribute to the tendency of concrete to
spall in fire. For example, an influence of the aggregates cannot be ruled
out. Probably this has more to do with intrinsic porosity than with the
kind of material involved. For example, spalling was observed in very
moist mortars made of lightweight aggregates. This may be due to the fact
that on rapid heating moisture stored in the aggregate cannot escape rapidly
enough through the more compact cement paste.
It is probable also that the tendency to spall is influenced by the
design of the cross-section and by the mechanical stress during the fire. A
preference towards spalling is noted, for example, in the slender stems of
I-beams and on the under sides of the cross slabs of T-beams. In these
members strong heating produces very high compressive stresses owing to the
inhibited expansion, and these stresses cannot be absorbed even by concretes
of very high strength. More precise investigations should be carried out
and published as quickly as possible into the causes of spalling, which
-28-
certainly does not occur on all occasions.
3.6 Columns
3.61 steel reinforced concrete columns
Whereas in steel reinforced concrete beams and floors the reinforce-
ments are generally subjected to tensile stress, in the case of reinforced
concrete columns the steel is stressed in compression. The compressive
strength of steel, i.e. the compressive yield point, is also reduced under
the influence of high temperatures, so that in the presence of fire there is
a greater danger of buckling.
In earlier years, extensive series of fire tests were carried out
C 4 8 - ~ O )
on steel reinforced concrete columns ). The fundamental tests of
Ingberg and his co-workers are still indicative. At that time concrete
quality grades were still comparatively low, the column cross-sections were
large and the reinforcement ratios were low. Very high fire-resistance times
were attained with plastered columns where the plaster was furnished with a
light wire-mesh inlay, which prevented the plaster from dropping off in the
course of the fire.
More recently there has been a strong tendency in construction
towards reducing the dimensions of individual structural members. In par-
ticular, architects have been calling for ever slenderer columns in fafade.
This has meant the employment of higher concrete grades and larger pro-
portions of reinforcement.
With the advance of prefabricated construction a desire has arisen
to use uncoated columns with high fire-resistance time, since subsequent
coating tends to nUllify the economic advantages of prefabricated construc-
tion. Therefore, new series of fire tests have been conducted recently at
various places with slender uncoated, and in some cases heavily reinforced
columns
C34,Sl-SS).
In addition to depending on the cross-section size, the resistance of
a steel reinforced concrete column is influenced primarily by effectiveness
of the covering over the steel. Owing to the severe compressive stresses in
the boundary zone, this covering is more severely stressed at the beginning
of the fire in columns than in other structural members. If the concrete
or coating falls off prematurely, then even thick columns will not attain
long fire-resistance times. The most important measures for raising the
capacity for resistance of steel concrete columns to fire lies in holding the
-29-
covering, whether of concrete or plaster, in place while it is being
attacked by the fire. This is best done by inlaying a light wire mesh
between the reinforcement and the surface of the concrete. When this is
done, constant heating of the column, and hence a more even course of the
fire test, is obtained. The following observation may be made in here. With
the use of certain aggregate materials, e.g. limestone or blast furnace
slag fragments, the concrete covering is less liable to falloff. In such
cases, the wire mesh inlay can sometimes be dispensed with.
The bearing capacity of a compressively stressed column is described
approximately by the following addition law:
p
collapse = K
b
. F
b
+ 0st . Fe
where:
K
b
cube strength of the concrete
F
b
cross section of the concrete
o = tensile yield limit of
the steel
st
F = steel cross-section
e
In principle these conditions also apply in the course of heating,
taking into account, obviously, the altered properties of the material.
From the equation the following conclusions may be drawn:
With increasing concrete cross-section, given to equal reinforcement
ratio, the fire-resistance time has increased, because complete heating
of a larger cross-section up to the critical temperatures must take longer.
However, increasing the reinforcement percentage, i.e. increasing
Fe' under otherwise similar conditions has not so great an opposite effect,
since the steel reinforcements of a heavily reinforced column heat up just
about as quickly to the critical temperature as they do in a lightly 're-
inforced column.
The time until failure is also influenced by the size of the
applied load. No data are available on the influence of the degree of
restraint and the slenderness on the fire-resistance time. It can, however,
be assumed that the test results would be greatly influenced.
Figure 9 shows the results of recent fire tests with prefabricated
columns carried out under the auspices of the Federal Association of the
Concrete Block Industry(53). In this, columns 3.60 m long with the following
cross-sections were tested:
F 15 cm x 20 cm = 300 cm
2
F 15 cm x 24 cm 360 cm
2
-30-
F 20 cm x 20 cm = 400 cm
2
F 24 cm x 30 cm 720 cm
2
For some of the columns limestone was used as the aggregate, and
for others a quartzitic material. Figure 40 shows the column cross-sections.
Figure 41 shows a column after the fire test and the quenching water test.
The inlaid wire mesh is clearly shown.
Figure 39 shows a definite increase of the fire-resistance time
with increasing concrete cross-section. Striking, however, is the excellent
behaviour of thin columns, especially when limestone is used as the
aggregate. British fire tests showed a similar result on somewhat shorter
columns, and some of these are plotted in Figure 39 as well(3
4).
In
Figure 42 the results of the fire tests of the columns represented in
Figure 40 are compared with the results of more recent French fire tests(5
2).
Since in the latter the test arrangement (column length 2.30 m) differed,
the results are not directly comparable. Nevertheless it can again be
recognized that the fire-resistance times in the case of a French test,
as in the German results, are more or less proportional to the concrete
cross-section. In the French tests the reinforcement ratio ~ was substan-
tially decreased with increasing concrete cross-section, so that the applied
load on all columns was equal. Despite the radically altered reinforcement
component, the tendency is the same as in the test with constant reinforce-
ment component.
Another successful means of increasing the fire-resistance time of
columns was found to be the displacement of a substantial part of the re-
inforcement towards the interior of the column. With heavily loaded spiral
columns, which have a somewhat greater resistance than rectangular columns,
fire resistance times of more than 1 1/2 hour were obtained(55).
To sum up, the behaviour of steel reinforced concrete columns in
fire may be represented as follows:
1. The most important preventive means is to see that the covering
of the steel does not drop off. A wire mesh inlay will
accomplish this.
2. The choicE of aggregates has a very great influence. Limestone
concrete columns behave better than concretes with quartzitic
a g g r e ~ a t e s . Basalt and blast furnace s ~ a g s fall somewhere
between the above-mentioned groups.
3. Under the calculated permissible load columns of larger cross-
sectional dimensions fail later than slender columns. The
-31-
influence of the reinforcement ratio is slight compared with that
of the concrete cross-section.
4. Given suitable design and a good use of proper aggregates, high fire-
resistance times can be obtained even for very slender columns.
3.62 Unreinforced columns
From an earlier time we have the results of fire tests on unrein-
forced columns(SO). Figure 43 shows that very thick columns may fail faster
in a fire than columns of moderate dimensions. This observation has some-
times been made in real fires as well. It should be borne in mind, of
course, that the concrete grade of the tested column was very low.
The reason for this comparatively poor behaviour of the thick
columns may be that the share of the heated surface in the total cross-
section of the column is comparatively small. The interior of the column,
owing to lower temperatures does not then share the longitudinal elongation
of the surface, and heavy loads must be borne by the thin surface. In the
case of low-grade concretes, the bearing capacity in the boundary zone is
then quickly exceeded.
4. The Behaviour of Prestressed Concrete Structures in a Fire
4.1 General
Non-prestressed steel-reinforced concrete and prestressed concrete
have much in common from the standpoint of fire. The decisive factors as
far as the fire-resistance time of a structural member is concerned are its
cross-section and the heating of the steel reinforcements. Any means by
which this heating is delayed have a positive fact on the resistance of the
construction to fire. The heating of the steel reinforcements is affected
essentially by the size of the cross-section and the shape of the member, as
well as the covering of the steel.
Generally speaking, the dimensions of prestressed concrete
constructions are smaller than ordinary steel-reinforced member of equal
bearing capacity. This entails a smaller heat capacity and a more rapid
complete heating of the members, and has the effect of reducing the fire-
resistance time. On the other hand, the example of thin, heavily steel-
reinforced concrete columns dealt with above, where similar conditions
prevail, shows that suitable means are available in order to give even
-32-
slender' I on s a relati vely hi gh resistance to fire.
The main diffcrerlces between prestressed and ordinary steel-rein-
forced concrete f'r-o:n the [Joint of view of fire technology are as follows:
1. The reinforcement cods are of higher quality (cf. Figure 14).
The drop in high temperature yield point of the prestressed
reinforcement rods is greater than that of ordinary structural
steels for non-prestressed reinforcements.
2. The steel stresses are not proportional to the external moment.
Therefore a reduction of the external load is not so favourable
as in non-prestressed concrete.
3. Different dimensioning procedures are employed. In many in-
stances the actual steel stresses in non-prestressed concrete
are lower than the calculated stresses. This results in an extra
safety factor with respect to fire.
4.2 Prestressed concrete floors
Prestressed concrete floors are usually prefabricated parts. The
shapes employed are similar to those of non-prestressed prefabricated floors.
The different types may be used either with or without additional poured-in-
place concrete. The bearing parts of prestressed concrete floors are
generally produced in the prestressing bed with immediate bond.
The critical temperature of the reinforcement rods in the case of
prestressed concrete floors is somewhat lower than for steel-reinforced
concrete floors. The reason is that the ratio of the working stresses of the
steel in relation to the decrease in high temperature yield point is less
favourable than for ordinary steel-reinforced floors. Moreover, with
progressive heating the prestressing, owing to changes in the modulus of
elasticity of steel and concrete, and owing to the creep of the heated steel,
is gradually relaxed and finally disappears altogether. When this happens
the reinforcement then acts simply as a non-prestressed reinforcement. The
neutral axis changes its position. This results in severe overloading of
the compression zone of the concrete, and ultimately results in failure.
Characteristic of the behaviour of prestressed concrete floors in a fire is
a very marked more rapid increase in its deflection before fracture occurs.
This happens in non-prestressed floors, as well, of course, but is not so
readily recognizable.
Tn order to attain a maximum fire-resistance time, it is necessary,
-33-
for the above reasons, to delay the of the prestressed reinforcement
rods in the fire as long as possible. An effective way of doing this has
been found, for to be the application of an ancillary layer of
lightweight concrete below the floor beams. This increases
both the heat insulation and the adherence of the plaster, which is of
great importance in the attainment of a high fire-resistance time value.
In Table VI we represented two typical prestressed concrete floors
which satisfied the required fire-resistance time of 1 1/2 hours in Germany
("fire-resistant" construction)(37). Many other floor tests are described
. (52,56)
In .
A considerable improvement in the fire-resistance times of pre-
stressed concrete floors can be attained by means of special plasters (cf.
Section 3.31) and also by suspended membrane ceilings. Either of these
means increase the fire-resistance time up to 4 hours and more incertain
designs.
A slight increase in the fire-resistance time can also be obtained
by "over-dimensioning".
As already stated above, in prestressed concrete designs the external
moment is not proportional to the reinforcement stresses. Nevertheless,
the fire-resistance time of a floor under small load is somewhat increased,
because during relaxation of the prestress the instant at which tension
appears in the compressive tensile zone occurs somewhat later. There have
been exceptional instances, of course, where the failure of unloaded floors
occurred earlier than in similar floors subjected to a load. Presumably
this is because in the case of the unloaded floor the compressive stress
due to prestressing and thermal expansion on the side exposed to the fire
was too great, and hence premature splitting-off of the reinforcement
covering occurred.
To sum up, the following can be stated about the behaviour of pre-
stressed concrete prefabricated floors:
1. The decisive factors governing the fire-resistance time are the
dimensions of the cross-section and especially the thermal in-
sulation of the reinforcement covering.
2. The critical steel temperature is somewhat below that for non-
prestressed reinforced floors.
3. Suitable coats of plaster or suspended membrane ceilings can
result in very long fire-resistance times of 4 hours and more.
-34-
4. Most f i r e p r o o f i n ~ measures and modifications have effects
similar to those exerted on non-prestressed reinforced floor
construction.
4.3 Prestressed concrete beams
In recent years there have been numerous systematic experiments in
Great Britain(57), the Netherlands(4
7),
the U.S.A.(58,59,67) and in other
countries(60-63,69) on the behaviour of prestressed concrete beams in fires.
In these tests the following influences, among others, on the behaviour of
prestressed concrete beams were investigated.
1. Form and dimensions of beam cross-section
2. Covering of reinforcement rods with concrete
3. Nature of the bond
4. Kind of aggregate
5. Load
6. End restraint
7. Additional heat insulation of surface
8. Causes of spalling
Although agreement in all points is not found in the various test
reports, they nevertheless furnish very valuable information.
Since in the case of a beam the heat attacks from several sides, it
is heated through more rapidly than a slab which is exposed to the fire on
one side only. The smaller the beam cross-section, the less time it takes
to heat it through. For this reason, as in the case of steel-reinforced
concrete columns, the size of the cross-section has been found to be an
essential factor governing the fire-resistance time.
More difficult to recognize is the influence of the shape of the
cross-section. Rectangular and thick I-crass-sections behave very similarly.
With flat beam cross-sections, and the TT-cross-sections which are common
in America, it is less easy to eliminate the size of the cross-section as
a parameter for the fire-resistance time. Basically, however, for these
forms also it is true that the fire-resistance time increases with increasing
cross-section area. Table VII gives a survey of the cross-section forms of
steel-reinforced concrete beams investigated.
The rate of heating of the steel reinforcements depends not only on
the beam cross-section, but primarily also on the covering. The higher the
desired fire resistance, the greater must be the amount of concrete covering
-35-
of steel reinforcements. In Figure 44 we have the results of several tests
on steel-reinforced concrete beams with instant bond. The influence of the
beam cross-section and the reinforcement covering is clear. The minimum
steel coverings listed in Table VIII are obtained to ref. (47) for
the various fire-resistance times, taking into account the beam cross-
section. The values given are based on the assumption that the covering
will not falloff and that the concrete is of a technological quality such
that no spalling will occur. From a comparison of Figure 44 and Table VIII
it can be inferred that the of Table VIII already provide a certain
safety factor.
The values listed in Table VIII apply to compact cross-sections.
With very high, slender beams the covering must sometimes be still further
increased. It is very important, especially in the case of larger beams,
that the concrete covering cannot falloff during the fire. For this
reason longitudinal bar and lateral tie reinforcements must be applied.
In comparing beams with instant and delayed bond, similar results,
on the whole, were obtained(47). With small coverings, beams prestressed
after pouring often showed somewhat better behaviour. This is because the
high cement injection mortar in the prestressing channels has poor heat
conductivity (no aggregates) and a substantially higher specific water
content than the concrete. This water content delays the heating (cf.
Section 2.24). To be sure, the high content can also have a deleterious
effect if the vapour pressure becomes too great and causes concrete to spall.
Escape routes for the water vapour should therefore be provided.
The investigation of a lightweight concrete of bloating clay with
about 2/3 the bulk density of the heavy concrete investigated for comparison,
resulted in about 20% longer fire-resistance times(59). The author is un-
aware of many systematic investigations of the influence of other aggregates
such as limestone or blast-furnace slag, which was very effective in the
case of columns.
Reducing the external loading has a positive effect. However, for
the reasons explained in Section 4.1 this influence is not nearly so great
as in unreinforced concrete(47).
As in the case of non-prestressed reinforced structural members,
we do not have adequate information, either, about the influence of end
restraint on the fire-resistance time of prestressed concrete beams although
in practice such restraint is almost always present. In British tests(57),
where the longitudinal expansion was prevented, an increase of fire-resist-
-36-
ance time was found for small beam dimensions, but a decrease for large ones.
According to American investigations(58) the effect of fixing depends
decisively on the magnitude of the "restraininR; force". This influence is
represented systematically in Figure 45. AccordinR; to this curve the
shortest fire-resistance times are obtained for total end restraint and
simple supports. Between these two extremes the bearing capacity under
fire stress can be prolonged in such a way that, for example, in the case
of space separating structural members of prestressed concrete the increase
in temperature on the side away from the fire, not the collapse of the
member, becomes the determining factor for the resistance time.
Obviously prestressed concrete beams can be very well protected
against attack by fire by means of an additionally applied heat insulation
layer, e.g. of vermiculite concrete. According to Ashton and Bate, encasing
a beam in vermiculite concrete 25 mm thick increases the fire-resistance
time by more than 2 1/2 hours. American tests show(57) that the fire-
resistance time of a TT-beam is doubled by spraying on a 12 mm coating of
vermiculite and tripled by spraying a 25 mm thick coating.
5. The Behaviour of Other Concrete Parts and Construction in a Fire
5.1 Chimneys of lightweight concrete pipe sections
Unlike the structural members hitherto described, chimneys are
under stress from high temperatures not only in the catastrophic case, but
also in the course of normal operation. To be sure, the temperatures of the
combustion gases, which are brought through a flue, are considerably lower
than those that occur in a severe fire. In domestic heating plants
temperatures between 200 and 400°C can be expected.
In addition to these "normal" temperature stresses, however, still
higher temperatures can occur in chimneys. This is the case, for example,
in chimney fires and so-called "burning-out". Certain fuels tend to deposit
considerable quantities of shining soot in t ~ e chimney. This soot under
suitable conditions can be ignited. This can happen either involuntarily,
or can be brought about deliberately by the chimney sweep or the fire
department and kept under control. When chimneys are being burned-out,
temperatures of 10000C and more occur(64). Different views are entertained
on the effective duration of these high temperatures(6
5).
They probably
jepend very strongly on the specific conditions, especially of the fuels
-37-
employed. In any case, conditions during are very similar to
those arising in a fire.
A number of requirements must therefore be laid down for chimneys
from the point of view of fire protection:
1. In normal operation the outside should not get too warm, so that
undesirable phenomena such as the discolouration of tapestries,
etc., will not occur.
2. They must be sufficiently gas tight to prevent any combustion
gases from leaking into adjoining rooms.
3. In a chimney fire, the temperatures of the outside of the
chimney must not exceed the limits laid down for space separating
structures (cf. Section 3.1), so that no ignition by transfer
will occur in adjoining rooms. Hence any cracks that may occur
must not exceed a certain width.
4. Chimneys must have adequate mechanical strength, because they are
subject to mechanical stress both during fires and when being
swept.
5. They must be sufficiently stable.
The above-mentioned requirements are determined by tests which differ
in execution from the normal fire tests described in Section 3.1. Figure 46
shows the temperature-time curves for heating and burning-out tests according
to DIN 18160, sheet 6 "heating plants-testing principles for domestic
chimneys". First a specimen is subjected to a heating test, and then, after
cooling to a burn-out test.
In recent years chimney pipe sections of lightweight concrete have
proved very effective. Systematic tests have been carried out by Seekamp
and Mohler(65). Many chimneys made of such pipe sections have also been
tested on behalf of individual companies.
As an aggregate for lightweight concrete chimney pipe sections,
preferably blast furnace pumice stone slag, crushed brick, bloating clay and
broken, porous lava slags are employed. In order to meet the contradictory
requirements of good heat insulation and adequate gas tightness, a narrowly
restricted range of bulk densites, depending on the material, must be
observed.
For small cross-sections generally one-piece units are employed
which may be of either solid wall or cellular design. For larger cross-
sections, multiple-shell units have been found more effective. Where
chimneys have to be burned-out fairly often it has been found useful
-38-
to incorporate a thin reinforcement of annealed wire in the chimney units.
Figure 47 shows examples of several types of chimney pipe units.
5.2 Glass block walls
Glass block walls occupy a special place in the field of wall
construction. In a fire these must remain space s e a l i n g ~ No requirements
are stated for the rise of temperature on the side away from the fire.
Figure 48 shows a glass block wall in the course of a fire. This wall has
satisfied the requirements laid down in the German regulation, whereby glass
bricks must withstand standard fire for one hour and then withstand the
quench water tests. The deformation of the wall, which after cooling is
restored to its original shape, is clearly evident in this picture.
5.3 Roofing
Concrete roofing materials consist predominantly of:
lightweight concrete slabs,
asbestos-cement slabs, or
concrete roofing tiles.
For the case of lightweight concrete slabs, what has been said in
Section 3.3 again applies. The latter two products constitute roof covering
materials in the narrower sense. Asbestos cement slabs and concrete roofing
tiles are among the so-called "hard roof coverings". Both types are
sufficiently resistant to air-borne fire and radiant heat, i.e. they
protect a building from the effects of neighbouring structures on fire.
Figure 49 shows a timber frame building which is veneered on the gable side
with asbestos cement slabs. The building is completely undamaged, although
a neighbouring building has been burned to the ground.
6. Mathematical Determination of the Fire Resistance Time
Attempts were made at an early date to determine the fire resistance
of construction mathematically in advance. For example, Busch(9) attempted
to calculate the time taken for structural members to heat through in a fire
with the aid of the familiar general Fourier differential equation for heat
transfer, and from this to determine the fire-resistance time.
The values thus obtained and the heating through curves are very
interesting, to be sure, and give many indications and suggestions, but
they describe the actual conditions in many cases unsatisfactorily. This
-39-
is mainly because the material constants change radically with ttle
temperature and generally they are difficult to take into account. Also
the conditions under which a test is run can vary greatly, so that for
example the moisture content may have a decisive effect on the behaviour
during a fire.
Recently attempts have been made to deal with the behaviour of
constructions analytically not on the basis of Jrathematical theorems, but to
draw general conclusions from the results of fire tests on various structural
members. In the Netherlands, for example, the rate of heating of steel
reinforcements has been suggested(4
7
) as a basic parameter for the behaviour
of reinforced concrete structures. It can be assumed that with detailed
analysis of the test material at present on hand, a number of other essential
influence factors can be eliminated. This would apply, for example, to the
size and shape of cross-sections, the static system, of the kind of aggre-
gates, etc.
On the basis of these magnitudes, a sufficiently accurate advance
estimate of the fire-resistance time should be attainable. At the same
time, it should always be borne in mind that the required fire-resistance
times, like the variation of the fire with respect to time as assumed in
the standards (temperature-time curves), are only conventions. For the
behaviour of a structure in a real fire, it is of minor importance whether
a given structural member withstands the standard fire, e.g. for 1 hour
and 50 minutes or for 2 hours and 10 minutes. The order of magnitude is
what matters. In any event it makes little sense to the author if a large
number of constructions very similar to each other have to be tested in
extensive experiments in order to satisfy a standard. This is precisely
what has happened in recent years in the field of concrete products and
prefabricated concrete parts.
7. The Behaviour of Concrete Products and Prefabricated Concrete Parts in
Real Fires
In the foregoing sections we have dealt mainly with the behaviour
of the various concrete constructions in the fire test under standardized
condjtions. Only through experiments under strictly defined conditions
was it possible to investigate systematically the influences on the
ppsistnnce of structural members. However, it is the behaviour of construct-
-40-
ions in a real fire which in the last analysis really counts. To what
extent do the results of fire tests accord with experience from real fires?
From the very outset it cannot be expected that detailed findings,
for example, the propagatioll of a temperature through a construction, will
be in exact agreement in fire tests and real fires. A comparison must
rather be made primarily in terms of the general behaviour. Considered this
way there is far-reaching agreement between tests and actual fire results.
To be sure, the elimination of influence factors is more difficult with
respect to real fires than for fire tests. On the other hand, in fire
disasters, a number of characteristics become evident which in tests could
not be so clearly observed.
Generally it may be stated that almost all measures which increase
the bearing capacity of a construction under normal temperature conditions
are even more effective against stresses due to fire. This can be realized
particularly after fires in older buildings from the early days of concrete
construction, in which the present-day building code had not yet been
applied.
Like the fire tests, practical experience also shows that the
covering of the reinforcements is of particular importance in all steel-
reinforced concrete structural members. By investigations on burned
buildings it has been established(66), that the thickness of the reinforce-
ment covering is less important than that it be prevented from spalling
during the course of a fire. To ensure this, in beams and columns suffi-
ciently compact longitudinal bar and lateral tire reinforcements are re-
quired. Figure 50 shows that if the stirrup spacing is too wide the rein-
forcement covering falls off, whereas if the tie spacing is more compact it
stays on. Also, the diameter of the ties, especially in the case of
columns, should not be too small.
The falling-off of the covering is also favoured by a narrowly
spaced supporting reinforcement. The great differences of stress occurring
in the concrete in a fire are especially critical at points where the
concrete cross-section is reduced, i.e. at and between the steel reinforce-
ments. The observance of the minimum distances between reinforcement rods
laid down in steel-reinforced concrete standards is therefore of special
importance from a fireproofing point of view.
It has been found important, especially in older buildings, to have
adequate thrust reinforcement and adequate anchoring of the reinforcements,
since otherwise the considerable deformation due to the high temperature
stresses cannot be absorbed.
Furthermore, the bearing capacity of concrete constructions in
fires in a large number of cases has been found in practice to be greater
than in the fire test. This applies particularly to constructions which
were tested as statistically determinate bearing structures, but which
in practice are statically indeterminate. This is the case for numerous
prefabricated concrete constructions, especially prefabricated floors.
The latter are almost always tested as monoaxially restrained. In practice,
however, there is a certain lateral support as well. In fire tests, where
in general free lateral movement is permitted, it has been found that slight
lateral support or restraint produces a considerable increase in the fire-
resistance time.
One effect which is not covered at all in fire tests is the size of
the building. In large buildings it has been observed after fires that such
large mutual displacements of entire sections of a structure took place
as a result of great changes of shape, that the individual structural
elements were no longer able to absorb these displacements(68). For this
reason, in all buildings exposed to the danger of fire there must be an
adequate disposition of expansion joints.
To sum up it may be stated that for concrete, steel-reinforced
concrete and prestressed concrete constructions there is good agreement be-
tween behaviour in a fire and behaviour in a fire test.
Failure to take into account generally benign construction principles
which are of great importance in the case of a fire, has an unfavourable
result. More favourable in many cases, on the other hand, are certain
concealed static reserves which do not show up in the test.
8. Summary
Concrete and reinforced concrete construction has existed for over
a century. Throughout this period it has been found over and over again
during fires that concrete is a building material that is outstandingly
resistant to fire and is, in this respect, rivalled by hardly any other
material. It has the advantage over combustile materials of being incombus-
tible, and it compares favourably with metals by virtue of its relatively
low thermal conductivity, its large mass and its high specific heat.
Despite these favourable conditions, concrete and reinforced concrete
structures are not able to resist fire indefinitely either. During a fire
-42-
the structural members affected are gradually heated through and through.
Depending on the dimensions of a member, this heating process proceeds
at a higher or lower rate. With progressive heating, changes occur in the
material properties of the concrete and reinforcing steel. In addition,
the distribution of forces within the cross-section and sometimes also the
stability conditions are affected. Since the penetration of the heat occurs
more slowly in members having larger cross-sectional dimensions, these
dimensions are, generally speaking, the most important factor affecting the
behaviour of concrete members in fire. Furthermore, in the case of reinfor-
ced structures, the fire resistance is very largely dependent upon the
protection of the reinforcement against excessive heating. The thermal
insulation afforded by the concrete cover to the steel can be improved by
plastering or by special selection of the aggregates employed.
To summarise, it can be said that it is nowadays possible, by means
of suitable technological and constructional precautions, to build con-
crete, reinforced concrete and prestressed concrete structures for any
desired fire-resistance period within the limits encountered in actual
practice. By taking advantage of the knowledge gained in fire tests and in
fire damage to buildings it is possible also to obtain considerable f i r ~
resistance periods in precast concrete construction with the slender mem-
bers favoured by that method of construction.
The great spread of concrete and reinforced concrete construction
over the past hundred years has occurred not least because - in addition
to other advantages - these materials are superior to other building
materials in the event of fire. It can safely be assumed that in the
future, too, this superiority will be retained with respect to new building
materials and methods.
9. References
-43-
(II S. D. Hami lto'n. "A Short History uf the St.ructural Fire
Protection of Building!'.", London J9SS, Herr Mayesly's
Stationary,
(2) l\.f. Gnr y ...Drandprobcn an gtscnbctonbauten", Deutscher
AusschuU fur Eiscnbcton Heft u, und 41. Berlin 1911,
1916. 1!1l8.
[3] ..vcrstencrunuswtrtscha rt", 17. Jahr.l!ang Nr. 6, 15. Man 1962.
(4) ..g ta t.istisch cs Jahrbuch rur die Bunllcsrepublik Deutsch·
lnnd 1962", W. Kohlhammer, g'tuttunr-t.
(5) K. Enrtcll , ..v c rsucbc tiber Lannen- und
von Bc tonzusch laustof Icn und z crncntrnortctn ur.tcr Ein-
wu-kung von Temperaturen hi:; 1200
0
C",
Dcutscher AU5...chuB rur Slahlbelon, Heft 60, Berlin 1929.
Wilhelm }':rnsl & Sohn.
161 K. D. Nckrassow, ..HltzebeslancHger Belon", Bauvcrlag
GmbH, Wlesbaden 19(i2.
17} H. l'hillco, "Some Physical Propcrt.ics or Concrctc at High
'pompcr-atu rcs", Journal of the Amertcan Concrete Institute,
April 1958.
[8) R. Grijn und Bcckmann, "Verhaltcn dcs Bctons in nonercn
Tempcraturen", Archiv tur Et!icnhUttenwescn, 1930.
(9) H. Dusch, ..Feuercinwirkung aut nicht brennbare Bausfoffe
und DaukonstrukUonen", zemcntverrag Gn:bH, Berlin.
[10) S. ,:Ober <las Verhallen des Betons und Elsen-
betons Irn Feuer und die Ausbtldung von Dehnungsfugen
Im Eisenbetonbau", Diss. T. H. Braunschwe ig, 1935.
(1I] H. L. Malhotra, "The Elleet of Temperature on the Com-
pressive strength of Concrete", Mag of Corrcr, Res. Vol 8,
No. 23.
(12) K. auue, "Heat-Resistant Concrete", Civil Engineering and
Public Works Review, Sept./Okt. 1962.
(13) Lehmann und G. Mall.ig. "Uber die HeiBdrllckfestigkeit von
Beton", Tonindustrle-Zeitung und Keramtsehe Rundschau
Nr. 84 (1960).
(14) A. Mehmel, ..trntcrsuchungen uber <len EinfluB von Tem-
peraturcn uber lOOn C auf Druekfestlgkeit von Beton",
Kurzber. Bautorscn. 4 (1963) Nr. 5, S, 78-79,
[15J L. Ludern, "Fcuel'beton auf POl"tlandzement-Basls", zemenr-
Kalk-Glps, 1959.
(16) L. Ludera, Fire Research 1957, Herr Mayest)·,s Stationary
Office. London 1958.
(1'JJ -t,e comportement du be-lon entre 80 et 300
0
C., S. R. T.
Batlr No. 102, April 1961, Paris.
(18) Woolson, "A.S.T.M. Proceedings", Vol. 5, 1907.
[l9) C. R. Cruz "An Optical Method Cor Determining the Elastic
Constants of Concrete", Journal of the PCA Research and
Development Laboratories May 1962.
(20) E. Raisch, "Die Wlirmeleitflihigkelt von Beton In Abhanglg-
keit vom Raurngewicht und Feuchllgkellsgrad", Gesundheits-
inuenieur, Sonderheft, 1930.
[21] O. Kriseher, "Die wlssenschafllichen Grundlagen der Trock-
nungstechntk", Springer Verlag, Berlin, 1956.
(22) Personltche Mitteilung der Firma Deutsche Siporex GmbH,
Essen.
(23) Personllche Mitteilung des Fachverbandes tur Hochofen-
schlacke.
(24) Htitte, Taschenbuch rur Eisenhtittenieute, 5. Aurlage, Ber-
lin, 1961.
[25) Kristen. Hermann und Wedler, "Brandversuche mit belaste...
ten Eisenbetonbautellell und Stelnelsendeck:en", D.A.t.St.B.,
Heft 89, ncrun, 1938.
[26] Dannenberg. Deutschmann und Melchior, "V,tarmzerreIBver-
suche mit Sparmatahleri", D.A.f.St.B., Heft 122, Wilhelm
Ernst & Sohri, Berlin 19.7.
(27] Ros und Eichinger, .Fesllgkeitselgenst·haften del' St"h1e bel
hohen Temperaturen", EMPA, Bericht Nr. 138, Zurich 1941.
[28) Ros und Eichinger, Werkstoffhandbuch Stahl und Elsen, 3.
Auflage.
(29] H. Seekamp und W. Becker, .Erhohung der Feuerwlder-
standsdauer von Stah1betonrippendecken durch Verwen-
dung von Holzwolle-Lelchtbetonplatten", Bautechnlk 1/1962.
(30) H. Seekamp, "Problemallk und Systematll( des Verfahrens
zur Prtilung des vernaucns von Holz und Holzwerkstolfen
im Feuer", Holz ats Roh- untl Werkstoll, 14. Heft 8.
(31) H. D. Forster, E. R. PInkston, S. H. IngoHg, ".'Ire Resist-
ance of Walls Lightweight Aggregate Concrete Masonry
Units" National Bureau of Standards BMS 117, Washington,
.
[32) Schlyter, Meddelande 50, "PrUfung der Feuerbestiindlgkelt
von Baumaterlal", Stockholm, 1931.
[33J C. A. Menzel. "Tests of the fire resistance and thermal pro-
perties of sottd concrete" Proceedings ASTM, Vol 43 (1943)
(34] N. Davey und L. A. Ashton, Investtgauons on Building Fires
Part. V. "Fire on Structural Elements" National Building
Studies. Research Paper No. 12
Her Mayesty's Stationary Offlce, London 1953
[35J G. Blunk, Th. Kristen, Strey, Wallo, ,.Baulicher Feuerschutz.'·,
Fortschritte und Forschungen Irn Bauwesen, Reihe D. Heft 27,
Stuttgart 1957.
(36) Per,tlnllche Milteilunt: des Verbandes RheinischeI' Blmsbau-
storrwerke e. V., Neuwlcd.
[:i7] Th. Kristen, P. Bornemann, "Das Verhaltcn von Dccken-
konstruktioncn aus Stahlbetonfcrtig-tcllcn bet Brandvenu-
chen", Bau u. Baulndustrle Helt 2211959, DUsseldorf.
(38J Persorrltehe Mitteilung dcr Firma Verelnlgtc Baustolfwerke
Bodcnwerder GmbH.
[391 Perstlnllche Mitteilung del' Firma Lenz & Co .. Munchen.
[40) Perstlnllche Mlttelulng del' Firma Rhelnbau GmbH.
Omnla-Decken-Organlsation, Wlesbaden.
-44-
(-{IJ Mitteilung dcr Firma perrckra-naustorre Miescn
.& Co.. Gl'lscnkirehcn. .
[42] J. P. l-'nrkl c r, " de au feu de Poutrcs et
pfanchc rs «. Cnh ict-s du C.S.T.B. No. 44, Junl 1960
[UI TIL Krislcll, Ir.-J. Wicrig...tjntorsuehunucu uoer unatj nstlue
bci Brandvcrxuchen an Deckcn aus nouen-
ctnandcrttcucndcn Fertigbalken oder Platten", wird dcm-
nuchst verorrcnutcnt.
(HI Th. Kristcn. H.-J. •..Das Verhalten von Deckcnkon-
stru k t i on cn im Feuer bel Brandausbreitung von oben nach
unten. Betonstein-Zeltung 5/)959.
(45) r-crsonucne Mitteilung der Firma Betonwerk Karl BUrkle,
Stuturart.
[46] r'crsonucne der Firma Heumar, nlpl.-Ing. Knoll
& Co .• Koln.
(47) "Brandproevcn op vooruespannen Betonliggers", 13. Raport
C.U.R., Amsterdam
{4B] Ingbel'J.l: Griffin, Robinson Wilson, "Fire Tests or Building
Columns", Technical Papcr Nr, IBD Burcau of Standards
\Vashington 1921.
(49j Hill, lngberg, ..Fire Resistance of Concrete Columns". Tech-
nical Paper 272, Bureau of Standards, wasntngton 1925.
(50) Schulzc-Wcdter-. "Brandversuehe nut belasteten Eisenbeton-
bautcttcn", Tell II. Shulcn, Deutschcr AusschuB rur Stahl-
beton, Heft 92, Bcrf ln 1939.
(51) H. Seckamp. ..BI'andversuche mit stark bcwehrt en Stahl-
betonsaulcn", Deutscher Ausscnun fUr Stahlbeton, Heft 132,
Berlin 1959, Wilhelm ErnSl & sonn.
(52) J. P. Fackler,« Essaisde resistance au i'cud'elelnents de Con-
struction (Be series)», Cahiers du Centre Scientlfique ct Tech-
nique du Batiment No, H, Juni 1960 et No 49, April 1961.
(53] r'ersonttche Mitteilung des Bundesverbandes der BeLonsleln-
iuduatr-le.
L. A. Ashton and P.M.T. Smart, "Sponsored Fire-Resistance
"rests on St.ructur-at J:::lements", Her Stationary
Office isso.
(55) H. HcuCers, ,.Drandversuche an sehlankcn, stark bewehrten
Stahlbctonsaulen hoher Betongtite", Beton, Herslellung,
VC1'wcndunl: 13, (1963), Heft 5.
156) A. W. Hill, L. A. Ashton, The Fire Resist.nco: of Prestressed
Concrete Civil Engineering an Publlc Works Review Vol 52,
No 617, November 1957.
(57J L. A. Ashton und C. C. C. Bat", "The Fire-Resistance 01
t-rcs trcsscu Concrete Beams", JUst. or Civil Enginet-rs, Lon...
don 1960.
15B). A. H. GustaCclTo and C. C. Carlson. "An Interpretation of
Rcsut ts of Fire Tests or rrrcs tresscu Concrete nUilding' Com-
ponents", PCI Journal, Oktober l!.l6'l.
1591 C. C. C;lrboll an P. J. Tatm"lllll. "The Ncw tjcam Frn-nace at
PCA and Sornc Experience Gained frorn Its Usc", PCA BuI ..
tcttns 142, Skokie 1961.
{OOl H, A. Bnk kc. "Brann{orsk mcd vanug arrner-tc of for-spent
armcrte betonjetkcr", Norues nranntekntske taborntortum,
R:lpport Nr. J, 1,957.
(G1] Th. Kristen, H.-J. "Der Jo:inCluU holler Temperaturen
au( Bautct le aus Spannbclon", Der B;illingenieur. 1/)000.
(62] Th. Kluz, "T'he Fire-Resistance of Prestressed Concrete:
Concrete & Constructional Engineering, July 1959.
(631 K. Kawagoe, "Fire TcsLo;; of Prestressed Concrete Slabs", Pro.
cccdmus of the Symposium on Prestressed Concrete and
Composite Beams, Japan Society or Civil Engineers, Novem..
bel' 1955.
(f4) Th. Kristen, "Tcmperatunnessungen belm Ausbrerinen von
Schornsteincn'\ warrncwtrtschart. 8. Jahrgang (1935), Heft 7/8.
(65) H. Seekamp, K. Mohler, ..Brandversuche an Hausschornstei..
nen", Berlin 1956, Wilhelm Ernst & Sohn.
(66) ]\.. 1. Hannemann und H. Thoms, ..Widerstands(ahigkeit von
Stahlbctonbauteilen und Stahlstef ndccken bei Brdnden".
Deutscher Ausschun fur Stahlbeton. Heft 132, Berlin 1959,
Wilhelm Ernst & Sohn.
(67J C. C. Carlson "Fire Rcslstance of Conerete", ACI - Publica-
tion SP-5, Detroit 1962.
[6B] C. Reiter "Das lo'euerrisiko des industriellen St.ahtbeton-
baus", ..Versicherungswrrtscnatt" lB. Jahrgang Nr. 2
-45-
Table I
Test conditions for the measurement of thermal
expansion of cement paste
Author Storage before Dimensions of Water-cement Age at time
test specimen ratio of test
Endell
7
days in water Prism 0.26 28/50 days
then air 18 x 18 x 100 mm
Nekrassow wet* cylinder
- 28 days
,0' 16 mm;
L = 100 mm
Philleo water prism 0.40 28 days
51 x 70 x 152 mm
* Testing at 110°C, prior to the drying.
-46-
Table II
la.
Temp. inc. on
side away from
Ref.
Wall cross- Material fire
I section dimen- Bulk density R
sions, cm
(air ary)
Test
6T
in o(
Compress.str.a Dr
time
IMax
Moist.cont. F (h)" Mean
1
I
2
I
3
I
4
I
5
I
6
I
7
---
I
-5t-

I

I
Pumice cement
1/2 38 54

boards
I 36 100 x 33 x 5 cm

R...... 650 ke/1II
3
1 44 54

V
1
a Dr not tested
1 1/2 60 74
t%
F = 12.7% by wt.
1 1/2
test
P- ern
,
fin-
lime cement ished
mortar
(S 5 ("
1. 5
plaster, :
Pumice cem. boards
1/2 21 24
olastered
1

R"",,650 ke/cm
2
1
29 34
2 3G
r:0 a Dr not tested 1 1/2 45 54
'"

'"'
F = 12.7% by wt.
1
F?
1
1/2
test
fin-
_ r..':: '-
ished
·Slftm ......
aerated
I

conc.
l%
slabs 1/2
'4
40
I
120 x 50 x 7.5clD
,
22

R • nO kg/1Il
3
1 54
e)

a Dr • 35 kg/c1ll
2
1
1/2 57
e)
F
:::
26% by wt.
1 1/2
test
fin-
I

ished
I
I
j oint ..,-
*)
inadequate
I
cemented formation of joint
-47-
Table II Continued
4 22
aerated conc.
blocks
50 x 25 x 7.5 cm
R = 550 Kg/m
3
a Dr = 2) kg/cm
2
F = JL2% by wt.
no fines conc.
crushed clay
brick 7/15 mm 1/2 2
5 35
kc;/m
3
8 20
R = 1330
46 kg/cm
2
1
1/2 36 45
a Dr =
1 1/2
test
F not tested fin-
ished
solid slag
conc. blocks
24 x 12 x 10 cm
1/2
6
35- = 1310 kg/m
3
1 1/2 6 9
J" Dr
=
66 kg;/cm
2
2 21 36
deter.
3 -42 70
not
3 test
finished
slag conc.
1/2
R
1/2 16 7 35 1 15
2 39 47
F
,
50 54
3
test
fin-
ished
a 34
-48-
Table II Continued
---
9 33
10 33
~ o -
heavy concrete
1/2
- -
-
~
f-
quart:;;;itic sand
,
""'44
- v./ /
3.nd c;ravel
2
""71
-
33
R = 2200 kg/m
3
,,
3 ""80
-
0 Dr = 310 kg/cm
2
v0
V ~ / ~ L
F' not testerj
....... 5
test
fin--
z-. 2DOk!}/m
3
ished
rJ/Z- 0.63
-49-
Table III
Increase of fire-resistance time of massive concrete
slabs owing to design modifications
Modification
Approximately 30% reduction of
steel stress
Increase of steel covering from
1 to 2 cm
1.5 cm plaster of strongly adhering
lime cement mortar
Special plasters with vermiculite,
perlite or sprayed asbestos base
applied underneath
Increase of concrete grade from
B 120 to B ,225
Increase of floor thickness
from 10 to 18 cm
Additional time before
collapse in %
approx. 40%
10 to 30%
50 to 100%
200 to 1000%
5 to 30%
approximately 45%
+'
(Ij
Behaviour in fire test
+' .c
00>:: +'
>:: Q) >:: cD Mean temp.
DC
'rl 8 ·rl
H 0 >:: I Q)
(Ij 8
------
Q) 0
Floor design Q) 8
8
Q) o
..000>,"'"
·rl o (Ij Q)
No. Ref. Dimensions in em >::(ljoo
-r-l >:: +'>,
H <+-t+,
+' ·rl U cD 0 H 'rl
------ ·rl .0 ..'>; 8 Q).o
>::8'018
<+-t ::J [fJ
+' rll >::+' [fJ 0 Q)
cD '---" >:: '0 '---" [fJ'O <+-t'O 'rl >:: 0, H
0.. Q)'rl Q) >:: Q) 'rl +,Q)Q)

+' 0...,...,
8cD Q 8 o:r.:H8 o:r.:o<+-t
1
I
2
I
3
I
4
I
5
I
6
I
7
I
8
Heavy cone.
--
• J/,.
t
e,...",3, 65m 1/2
-
305 48
\ I
I
(: J, // ' < <>/;('ij;(:;:X/ ,/
- 1
h
""15 475 90
-
simply
1 34
pY
?
l
l
h
545 94

supported
: =


-"
I
I steel

M,..",610mkg 1
h :22m
c o Ll a o s e
Cone. s tr. : WJ6 '260k9/
cm l
Heavy cone. massive slab
I '\ I
t "" 2, 70m
1/2
h
0,35 70 7

r
:/(/ , /';// r . ///:;;.,
....
. ."- '"
. ..
simply
1
h
0,65 133 23
2 38
I\L f5..L 15 J <:>.1\ steel. supported
\ ... lll.s
....
Gypsum board slab with fibre-
M,...." mkg 1 1/2
h
1,05 205 63
I
glass added Fire test concluded
Cone. str.
; IV.
a
: 528 kg/CIn l
I
I

-
>-3
ill
cr I
I-' Vl
(j) 0
I
H
<
15
91
29
50
5'
69
'4
125
2
1/2
4
Fire test concluded
t = 3,98 m
fixed all
sides
crossed
reinforce-
mlf..!270 mkg
(Feldmoment)

l"

( < : - '. , .
\
Conc. str.: W
so:28Dkglcm
2
neavv concrete
;'las
34 3
Fire test concluded
'/2 I 0,7
4 37
--33

e.. 4,00 m
simply
s uo o or-t ed
M_960 mkg
1/2 4,05
9,15
28
6'
82

I;:::
(])
Fire test concluded
5 39
f, = 4,00 m
simply
suppo:-oted
If.,,,,1250 lIlkC
1/2
1 1/2
',1
5,2
6,4
21'
'54
48'
'5
n
104
H
<:
o
o
::::;
IT
,-"
::::;
>=

I
Vl
f-'
I
6 37
poured-in place
r---
- 62,5 -
!,= 4,00 m
simply
supported
J'.1=unknown
';2
1/2
1,0
',6
5,EI
155
296
'92
29
45
55
1,5cm lime cern. plaster Fire tes
1
concluder
9 I 41
Cone. s tr. ;
I
Ul
fD
I

!OJ
o:
f--'
(lJ
H
<
o
o
::::;
(1"
f-".
::::;
>=
(lJ
c,
11
46
26
44
21
53
98
124 2,1
3,9'
1, 1
1,75
1,4
7,75
Fire test concluded
1/2
1/2
1/2
1/2
e=3,85 m
simply
sUDDorted
t = 3,85 m 11 1/2 I
0,5 47
simply 1
1 0,9 94
supported 1 1/2 1,3 146
I
13
3 3,3 365 30
M = 2120 mkPl
Fire test concluded
I I
IvI=unknown
(,= 4,00 III
s Lmp Ly
supported
= t 280 mke
.-.,
c."vs("Inj SIQbs
"5;I'"r.. "
--so
cQverinq;

e p!Q-5let'"
I'"rt;. mesh
/ /,
Vt:
0
// //// -
r' / _-,--/' / •
v '
\
lIJ I \
\
Steel
P"",red-;" pioL <:.
22
40
8
7
-53-
Table V
Influence of aggregates on the fire resistance time of steel-reinforced
concrete beams after ref. (28)
No.
1
Kind of aggregate
Alluvial sand and gravel
Fire-resistance time
hours : minutes
3h : 03m
2 like 1, but about 6 cm
crushed clay brick
concrete on bottom of
beam
3h 32m
3
Leighton Buzzard
sand + Torphin
Whinstone (limestone)
more than 4h
(test discontinued)
No. Ref. Floor desiq;n
dimensions in em
Span
l (0"1)
Bearin'6
Bending mOrrEn
at mid-bay
rl (mkg/m)
'='est
time
in h
Behaviour in fire test
Deflect-I Mean temp. in DC
ion at
id-bay ]at re- Iat side
opposite
ment fire
Reinforcement
above: 4 twisted
Ibelow: 9 steel rods 3x8mm sl 145/16
t; 4,OOal 1/2 2,25em 51
37
ime cement
plaster
I- 515 .1
simply
supported
K • 1640mkg/ml 1 1/2
7,5 em
15,0 em
72
70

PJ
0'
f-'
(j)
-<
H
I
\,]1
..r:::-
I
2 37
einforcement

lime cement
laster
light wt
t = 4,OOm
simply
sunported
K a
1/2
1 1/1
0,9 em
',7 em
2.3 em
( 93)
( 10:5>
( 172)
13
"H
41
-55-
Table VII
Shape of cross-
,
IRe/- I
di"lensions in C'Tl
No
. b
I
c
I
d .
47
I
10 15
!
I
g'
to to
I
1
I,,;
:<; .Q
- - -
i
I
J.
I '0
I
(;0
:"0 oJ
'6
I
i
."

2'
.0 0
• I
ito
to to
2
,1\
-c - - -
// I

/_, /. r
60 119 20.5
. _.J.
I

-'
1TI-:
25.( '2.5 G.'
1',4
,
r;l,
to to to to -
.
e'f-
In 78,2 )0.4 .0
r- 0 ..,
lfr
T
25.4 12,5 G. ,
" ,4
2.6
4 S..
to to to to tc
I, ---'--J ___Co
50,6 )0.4 12,7 22',8 5.2
----, t:
rill
120
'5
6
'0
11
,
';ll
I "IJ"
--
14O 18 11.5 14 14
-.jei-
r
- - --
0 ooj
J[f:
150 50
" .2
.,
6
150 40
".2
"
,l..-
I I
210 45 20 2"',\
_CII..,
7
6'

15 24 ',0 14,5
1L "
r°-,
Ijl
8
6'
: 1 ,
." ...
50 lH.5
4 I ,

II
I
-56-
Table VIII
Minimum prestressed reinforcement covering values after ref. (41)
100-200 200-500 500-1000 1000-2000 2000
Desired fire-
resistance time
hours
Required minimum re
1forcement
in cm for beam cross-sections
covering
of
1/2
1
1 1/2
2
3
4
2
"---------
2
3
2 2
5
4
3 3
6 4
3
6
5
8
10
Bild 1.. Brandschadensvorlouf in der Bundesrepublik Deutschland in den
Johren 1954-1961
Fig. 1. Diagram showing 'jre damage in the German Federal Republic
in the years 1954-1961.
Fig. 1. Degel, par incendies dans 10 Republiqu8 Allemande duront les
onn.e, 1954-1961
500;---"" "--
m]Gross social
product in 1959
'100 r- . 1 0 9 DM ._ .
:liDamage due to 1960
iHire in 10
6
DM
300
------
1956
1557 1958 .. . --- ..
1955
r:'
1551,
P
I....
200
--1'1'1 -- ",. .... ... .... I".
,
I
:1:
,
100
-
Ii
1
I
...
J
I
,
I
.
I
"'I
1:1
0
1551

5
I I
I
I
I, -- "--I---
j---
>=:
--1-
0
.r! 3
.. -
[/)
I
>=:
I
(Ij
,
0..

QJ
r1
(Ij
>::
H
QJ
.r:::
.j..)
0
H
(Ij
QJ
>=: -1
.r!
0 200 600 800

Bild 2. Lineare Warmeousdehnung verschiedener noth
(5). 1. Grauwacke, 2. Sondstcin, 3. Kolkstein, 4. Hochofenl'uck-
Ichlocke, S. BosolI, 6. Ziegelbruch
- - - - - Unsicherer Verlou' infolge Gasobgabo
Fig, 2. Linear thermal upanlion of various pggregate5 according 10 (51:
1. groywacke; 2. 10ndSlonej 3. limellonej 4. crulhed bla.l.furnace .Iagi
5. b05al.; 6. broken brick.
- - - - - uncertain due to evolution of gas
2. Dilatation lineaire par 10 chaleur de diver. agregalS lolan (5).
1. Aglamere., 2. Gr'" 3. Colcaire, 4. Lailier de haul fourneou en
morceau.., 5. Basolle. 6. Tuile. concanee.
- - - - - - Determination incerloine a CO".. d" degagemen' del
gal
-57-
800 1000 f200
Tlimperatvr 'C
600 200
o
rl
(1j
8
H
<1> -1
.c
E--l
%
c '2
o
·rl
UJ
s::: f1
(1j
0.
><:
<1>
800 1000 1;00
Temperalur ·C
600 1,00
!
i
200
-1
I I I
i, I I
"C-.J-i-r-1

1\............. I" .
\. I '"

'" ! --- J.--- I
i :
- ...-.. I -- - -1------ 1------
......1 ! r--I i
i I I I
rl
(1j
8
H
<1>
.c
E--l
Bild 3. Linoar. Warmoousdehnung 'ion Z.mentltein (Porllond:r.ement) b.i
.rstmaliger Erwarmung
------- nnch Endoll (5)
- - - - - - Abkiihlun
g
} ell N k (6)
_ _ __ Erworm"ng nO 0 fOllOW
_,_,_ Ph;lIoo (7)
Fig. 3. Linear thormal axpanlion of hardened cemont pOlte (Portland
cemont) on baing healed for tho ,jrst limo.
according '0 Endoll (5)
- - - - - - chooli,n
g
l according.o NQkroliow (6)
_ - - cooling
-.-.- Philion 171
Fig. 3. Dilatation lin60irc por 10 chaleur du cimon. durci (cimen.
land) chouffe pour 10 priCmier. foil
solon Endoll (5)
- - - - - - R.froidiu8ment } $ INk (6)
_ _ __ Echouffemen. 8 on 8 rauow
- • -- • -- Philloo (7)
Bild 4. Linear. Warmeausdehnung 'ion Zemen'stoin b.i de, aw.iten Er.
hil:rung {Erwarmung und Abki.ihlung} noch Nekrouow (6)
Fig. 4. linoar thermo I cIpanlion of hardened cement pOlle on boing
hooted for the second time (heating and cooling) according to N.luos-
lOW (61.
Fig. 4. Dilatation lincaire par 10 chaleur du cimen. dvrel, chauff' pour
"n. sec ende foil (echauffomenl of r.froidilsementl lelon N.,kallow (6)
--
s:
.lar"
(
If
v"-:
,/

----
\
\
\
"
\
\
"
J
c
0
·rl
UJ
>=:
2
(U
P,
><:
<1>
rl
(1j
8
H
0
<1>
.c
p
H
-1
(1j
<1>
s:::
·rl
-2
...::l
200 400 600 800 1000 1100
Tempe"alur ·C
Bild S. Lincc re Wormeousdchnung yon Mort.ln aus Porlland... men.
und vers chiedencn Zuschlogstoffen noch (51
Mischungsycrholtnis 1 : J : 0,67 noch Gcwichtsteilon
1. Rhcinkiesel, 2. Kolkstcin, 3. HochofensIljduchlaclr.o, 4. 80solt
- - - - - - Unsicherer Verlauf infolge Abgab. 'Ion COl
Fig. 5. lincar thcrmal expansion of molars mode with Portland cement
and various aggregates according to (S)
MiJC proportions 1 : J : 0.67 by weight
1. Rhine grayel; 2. limestone; 3. lump blal'·'urnace Ilag: 4. balalt
- - - - - - uncertain due to eYal"lian a' COl
Fig. S. Dilatation lineaire par la chaleur de mortiers on ciment Part.
land evcc diyors agregCJIs lelan (5)
Prapar'ion du melange 1: 3 : 0,67 en poi'dl.
1. Grayier du Rhin, 2. Calcaire, 3. laitier de haut 'aurneau en marceauit',
4. Basalte
- - - - - - Determination incerlaine 0 caule du degagement du C02
-58-
Bild 6. Druckf.sligkeit von Boton in Abhiingigkeit von der Erhi'zungs-
tempera'ur
Fig_ 6. ComprGllive strength of concrete 01 0 function of the heating
lemperalur•
Fig, 6. Resistance Q 10 compression du b.'on en fanction de 10 tem-
p.ra'ure d'.chauff.men.
---'-.-
; __ Tension
8 fa (2
in weeks
6
time
2 .(,
eating
fOIJlC--.......---.---------.....,
cCrl
+-' ct:
oO,,--j 10
S:::+-'
Q) s:::
t ,,--j 0
0
1-- .......--......----........ --"---.....
(/)
Q)
,..,
Bild 1. Oruckfc"jgkcit im Beton noch mehrwochigor Temp.raturb.aft.pru.
chung. Milchunguerholtnis 1 : 6 n. Gew.·T.i1.n
Ausgongsdruckfeltigkeit 246/kg/cm
l
Prufung in heiBem ZUI'and
_ _ _ _ _ _ Prufung nam vorherig.r Abkuhlunlll
o t.<: Erhitzung auf 100
0
C
=.; "-. Erhitzung ouf 300
0
C
Fig, 1. Compressive Itrength of concr.t. aft.r b.ing lubi.ctod to .1.·
vat.d temperotur. for leverol w.ek••
Mix proportionI 1 : 6 by weight.
Inilial compreuiv. strenglh 246 kg/cm
1•
teated in hot condition
_ _ _ _ _ _ hllted after cooling
o "'" healing to 100
0
C
:-·1 '-" heating to 300
0
C
Fig. 1. RGsiltanc. a 10 campr.slion dWi b.'on apr•• louminion • un •
•• mphlllhit. • ...".. r.."tlttn' UU'Ut'''... M"I.".. ,,' ."
,.elul. Itthluenc. 1,,11 alo D ta compr.llien 246 kg/e",1
Euoi a I'elot chaud
_____ - E"ai apr'l r.froidi...men.
o "'" Echauffem.nl it 100') C
-; Echowff.menl U JOO" C
1000 fZOO
Iemperatur ·C
800 600 '000 200
%
120
Q)
.c 100
+-'
c,...,
0 UJ 80
.c
*-1-'
bJ1
s::: s:::
60
,,--j Q)
H
.c+-,
*0
+-' UJ
00
s:::'U
Q)H
20
H 0
+-' 0
U)
+-'
0
0
rc
E;,. froid 400 kg/c
m1
E;', rwid 340 000 kg/cm
J
E;i frui,1 2010000 kg/c
m1
E ;', froitl 250 OO!) kg/cm!
E ;\ fruid 290000 kg/em
J
E
i
'
n ld
400,000 kg/cm1
E"
old
340.000 kg/cm'
E"'''d 240,000 kg/em'
E"lIltl 250,000 kg/cm
1
E•.
111

1
290,000 kg/c",z
10 module E du boton. ,ur
BelOn avec colcair.
Beton ovec diaba.
B.'on
lild I. EinfluO hCiherer Temperaturen auf den von Bclon
0----0 W,Z 0,40 E
ka lt
400000 kg/em' (19)
6 t:. W,Z 0,60 E
knit
340000 kg/em' (19)
-- -- -- Kolk".in-S.'on E
knll
240000 kg/em' (181
Ojobos-S.,on E,,,,it 250000 kg/em' (181
_ . - _ Sand.,.in·S.,on E
ka ll
290000 kg/em' (191
fig. I. Effect of elevol.d tempero'urol upo" the modulus of elouicity eE)
of coftcr.'.
0- 0
1\---6
w.e 0.40
w,e 0.60
- - - - - - limel.lone eencrete
dleb ese concrete
_ . _ . - I.andUone concrete
r;9, I.•"f1uence. do lemperaiurci .Iovee.
W:Z = eow/cimonl E
ka ll
= E il frui.1
o 0 W/Z 0,40
2.----t:. WiZ 0,60
500 bOO
Tcmperalur ·C
+00 300 200 100
-59--
densi tv 510kg;!,
o D.2 0,. 0.6 118 1,0
water-filled pore space
total pore space
o
Hcol
m·n··,
;'0

r<

>.,
.j..)
·rl 1/,
:>
·rl
.j..) t2
o
':::l
10
-c
c:
a
0,8
o
rl
0,6
m
8
H
0,.
<lJ
.c
I:-<
0,2
Bild 10. ;, aines Gasbotons in Abhongigkei. Von Fouche
Ilgkelt und Temporalur noth (:ll)
Fig. 10'; !hormol conductivity ). of an aerated concr.t. a' a function
of humtdlty and tomporat",re.
Fig. 10. Transmission de 10 chaleur ). d'un beton-goz en fonction d.
I"h",midi.e of d. 10 temperature.
1000 1400 1800 2200
Bulk density
600
I I7fF
I /
IT
/i
1/ .I
/
I /
/ /
.:
/
/"
,,/
/

/
l/
/
""
.//
v.,.,""
k/
I

i
1.0
0.8
116
o
2DO
>., 0
.j..)
·rl
:>
·rl
.j..)
o
::l
o
c
o
o
kca!
m./l·"C
1,4
8ild 9. Warmclcilzahlcn von Beton in Abhangigkeil von Raumgewicht
und Fcuchllgkcilsgcholl bci Roulnlemperolur nach (20\
_______ fcucht
- - - - - - lof ltrockcn
- . -- . - kiinsllich gelrocknol
Fig. 9. Thermal conductivity values of Concrete 01 a function 0' bulk
dcn:oity and moil'uN conlcnt at room ,empcroturo, according to (:ZO)
- mailt
- - - - - air dry
-- . _ . -- artificially dried
Fig. 9. Coefficient de transmission de 10 chaleur du belon en fonetion
do 10 denlil8 8' de 10 tcncur en humidit. a la temperoture ambian••
••Ion (20)
humide
Icche 0 I'air
lache or,ifici.llement
R - 506 ku!m' (22)
R = 1670 kg!m' 161
R =·1153 kg/m' (231
function of temperalure.
R 506 kg/m
l
1221
R 1670 kg/m' (6,
R - 1153 kg/m
l
(23)
o. a
Fig. 11. Thermal I';onduc.ivity of concrote
o 0 aeroted Concrete
6. 6 fire-cloy concrolo
X X foomed Ilog concrote
Bild 11. Warmclcilzohl von Beton in Abhongigkeit VOn der Tcmperatur
o 0 Go.beton R = 506 kg/m' (22)
6. 6. R = 1670 kg'ml (6)
-----, Hull.nb,m,b.'on R 1153 kg/ml (2J)
Fig. 11. Coofficient do transmission de 10 chaleur du baton en fonclian
de 10 tomporolurc
o 0 Bclon-go<
Bolon ovec Chamolto
X X Beton avec ponce do loi.ier
8
I
!
I
I
I
I
I
-:

--- ---
/
t.>
/ ...... r:::
7
....

I
0.)
a,I
cs
rl
m
8
H
<lJ
.C
I:-<
keol

0,
zoo
"00
600 600 fOOD
Temperotur °c
-60-
BOO fOOD ooo
Iemperotur °c f(J(J 100 JOIJ 'HIf) 5{)() 600 700 800
Tempera/ur 'C

V

I
I
I
Total water
1/
-cant. at begin-

ning of the
I +- /
u.;
--1:.-:

k-': !
I
/
I
/

UJ <l
UJ .c fO
o.p
rl
G-<
.p 05
.c
bOl>'!.
·rl
(]) s::: 0
0

QJ.p
G .c
JO
·rl b
O
O"ri
(]) QJ
0. ;>;15
UJ
Bild 12. Gewichhvcrlult yon Beton aUI kolkholtigom Sand und Kie' b.i
Erhitt:ung
ZcmenlgehClII Z = 440 kg/m1
Wauen:ementwert W/Z = 0,40
Fig. 12. Lou of .eight of concr.t. modo with coicorooul lond grawel
on healing.
Comen, (ontcnl C""" 440 kg/m
'
Water/cemont rotio W/C -- 0.40
12. Porte d. paidl du betan avec table •• Browier (oleair. lor.
de
te ne ur en ciment I -I..sO I.g:'m 1
Q"",1'r.lm .. nl \V,'Z :.•,)
BiJd 13. Lineor. Wormedohnung einel Stahlen mit 0,4'/, C {24}
fig. 13. Linear thermal expanlion of a ,teeI containing 0.4". C (24j
Fig. 13. Dilalatian lincaire d'un aeier avec 0.4'/. C (24)
Bild U. Warmltreckgrenze (0,2 Oehngrenze) deuhcher Betan- und Spann"
stahle
B 5, I <t> 8 mm
- . - - ... - . . .. Torstohl (/> 8 mm
- - - 51 160;180 <t> 5 mm, kaltgozagen
-- . - . - . - St 60/90 (/> 26 mm. wormgewo!z.
Fig. 14. Hot yield point (0.2". p,oof stren) of German reinforcing
Iteell ond prestressing steeh.
------- 5lru(lurol sleel B 51 I, 8 mm dia•
• • _ •• - - • - _.. Tor sle,)I, 8 mm dio.
- - -- 51 160/180. 5 mm dio., cold-drown
- . - . - . - 5, 60/90, 16 mm d;a.• ha'·,allod
Fig. 14. Limite de duclilile t. choud (0,2 limite d'elllienlibilite) d'acier.
allemands pour beton el belon precontroint',
B St I a 8 mm
- - - - - - - . - - . - oeier Tor " 8 mm
- - -- 51 160/180 " 5 mm, e.lire a 'roid
- • --. • - • - St 60/90 " 26 mm, lo""ni. a maud
500 SOO
Iernperat ur 'C
400 300 200 100
I
" 16Q/180
'\
\\
\
\
60190 \.
-"

-.-:--

-li.Stiiib
----
----I
............. 'II"
.- 8St. 1
r---..
-

UJ
UJ
(])

.p
UJ
G-<
150
0
0

0.
(\J
0
100
*!llmm
2
200
.p

·rl
o
0. 50
'd
rl
(])
__
-61-
'/,
50

c
M
-r-l
ctl
>-;
30
-I-'
UJ
OJ
20
-P
ctl
r=
.;::1
fO
+)
r-l

100 200 300 400 SOD 600
Temperatur ·C
Bild U. Bruchdchnung oines FluB"ahl.s in Abhiingig".i. von der Tem ..
peralur
Fig. 16. Ultimote strain of 0 mild ,teel 01 0 function of temperature.
Fig. U. EXlen,ibili'e jUlqu' Q 10 rupture d'uR Geier coul. en 'onetion
d. 10 temperolur•.
-------<,
'-,
-,
<,
,
'-, ,
!
-------_._--_.__.-
,
,
o SO
kg/
mm
'
200
UJ1S0
UJ
QJ
>-;
-I-'
UJ
<t--;
0
100
o
>-;
Q
100 200 300
Annealing
.00 500 600
temperature DC
8ild IS. O,2-0chnprobe deutscher SpannslCihle bei 20
U
C nod. vorhorigom
<il.hen (26)
- • -- • - Warmgewall.ler Siobsiohl
S, 60,90 von 5,2 mm cp
- - -- Kaltgczogcner Spanndraht
SI 1601180 von 5 mm cp
Fig. 15. 0.2'1. proof lest for German pres'reuing etccls 01 20
0
C ofler
prsviQI,IS annealing (26).
-. -.-- Hot-rolled bar, SI 60
190.
5.2 mm dia. .
- - - Cold-drawn prestressing wire, SI 160/180, 5 mm diD.
Fig. lS. 0,2 euai a I'extensibili.e d'ocien allcmondi pour bOlon precon-
trainls ci 20
v
C oprel i'edloullemenl ci rouge (26)
- • -- • - Geier en baguette lamine 0 chou'"
S, 60;90 de 5,2 mm a
- - -- Fil en ocier etire a ftoid pour belon procontrainl
SI 160/180 de 5 mm ..
I
.r!
-I-'
UJ
2
QJ
»
Cr-<+)
o 'r!
°1
tr:
;j
rl
;j
'd
o
100 200 300 400 500 600
-C
Bild 17. E.Modul van unsi'iziorlem Flu611ahl in Abhangigkoil von der
Tomporalur
Fig. 17. ModUlus of ola •• icily (E) of non-,iliceoul mild steel 01 a
function of temperature.
Fig_ 17. Modioli. E d'u" Dei", coul. lanl lilieG en fane lion d. 10
temperolur.
-62-
Gild 19. Tomperalurzcitkurwcn wcrschicdene, Land.,
- - -- Ocutsc.hland (-I
Nicdcrlande, Frankreich, GroBb,i'onnien, ISO
(Voroehlog)
USA
- -- - Schweiz
Schwodon
(eJ Di. deulsche Einheil ••emperalurzeitku""e wird d....noch.' de,
ISQ-Kurwe angepaBt
Fig. 19. Time-tempera1ur. curves of varioul coun'ri.,.
Gormany (.)
Holland, F.-one.. Greot Britoin. ISO (propOlol)
-. -. -. - USA
- - - Swit'lcrlond
Swedcn
(-) The German standard time.temp.rature curv. is SOon to be odop'.d
to the ISO curve
Fig. 19. Ligncs de temperolUre-IGmpl de diver, pOyl
- - - Allcmogne (-)
Pays Bo,. France, Anglete"e, ISO (Propo,ition),
USA
- -- -- Suisse

i:
1
ollvmonde aero bienl" odapt'e a
I

I i
I
,
!
I
I
I I
i
!
: i
!
!
i
60 120 fOO 21,/)
300 3611 420
480 Min
f
Z 3

5 6 7
B h
Duration of test
I
!
Ii!
Q)
H
·C
.,...,
1'-OD
4-t
s::
0 1100
Q)
U)
;:S fOOD
0
.C
4-t
800
0
P.
600
S
Q)
.p
.00
n
qj
s::
200
H
Q)
.p
s::
0
H
kcot
m·h··(
r,,",pualur 'C
80
Bild 18. WCirmelcitzahl von Siahien in AbhCingigkeit von de, Temp.rolur
0---0 .oino. Ei••n (18)
I:; I:; O,S", C (221
Fig. 18. Thermol conductivity of Iteels 01 0 function of '.mp.rotu,••
C----Q Pu.. i.on 1181
6-----1:; O.S", C (22)
Fig. 18. Coofficien, de tron,m'uion d. 10 chaleur d'Ge'." en fonetion
Jc 10 temperalure
D-----o Fo. pur (18)
.c:. I:; O,S", C 1221
-63-
Bild 20. Zurn Brundveesvch vorbereitete Wond au. gescha8hahen Leicht-
betotlplatten.
Fig. 20. Woll 01 storevhiqh lightweight concrete slabs prepared for
the lire te,t.
Fig. 20. Parol dessni pour 10 res istcnee au feu d'une hauteur d'u"
e.oge- en plaques de baton leger
em
20 r-------------..,.--;I"'""':I
rJl
ui
QJ

..':<:
o
.r!
..c:
+J
rll0
rl
cd

'07,5
QJ
>-i
.r!
::s
0'
QJ
0::5
1\ 2 4 8 h
Duration of fire
Bild 21. 8randkammer lur Deckenprulungen (Schemal
Fig.. 21. Fir. test chamber for Ih. 1.,ling of floor. (diagrammaticI
Fig. 21. ChoMbre pour I"euoi aux
Bild22. Erforderliche Dick. mOlliver B.tonwand. in Abhongigkeit von
BrClnddauer und Raumgewidat R 'ur .in. maximal. Temperafur.rhohung
yon l4DO C auf der dem Feuer abgekehrten S.ite
Fig. 22. Requilit. thickne.. of solid concrete walls as a function of fir.
duration and bulk density R for a maximum temperature ril. of 140
0
C
Oft the .ide remote from the fire.
Fig. 22. Epaisseur necessaire de paroi. mallives en beton en fonction
d. la durie du feu et de la den.it' R pour une augmenta.ion moximo
de la lemperalure de 140° C du e6le oppa" au leu
-64
Bild 24. Temperaturverluaf in einer 15 em dick.n Galbe.onwand
(R 900 kg/m
J
) im Narmenbrandversuch. .two 8 Pro... n'
Fig. 24. Temperatur. curve in a 15 em thick aerated concre'. wall (bulk
density 900 kg/mJI in the standard fire test. Moisture content appro•• 8'1,.
Fig. 24. Propogalion do la 'emperature danl une paroi de 1S em
d'epaineur en beton-ga)'; (R = 900 kg/mJ) a I'euai ou feu narmalise,
teneur en humidile environ 8
1
/ 1
SOD
250
fa;
o 7 II
Distance from the wall
side exposed to the fire
1000 --
·C
115/J
QJ
H
;:::l
.jJ
cd
H
QJ
0.
EO:
QJ
8
rl
rl ·C
cd
1{)(,(J

QJ
..c
800
.jJ
C
·rl
600
..c
.jJ
·rl
+00
QJ
H
;:::l
100
.jJ
cd
H
QJ
0
0.
5
EO:
QJ
side
8
Bild 23. remporalurverlouf in 5chwerbetanwandcn {R - 2300 kg/mJ) ver-
schi.denor Dicko im Narmenbrand nom V2, 1, 2 und 3 5.unden
Fig. 23. Temperature CUnt. in dense concrete walls of various thicknesse.,
as obtained in standard fire after Y2, 1, 2 and 3 hour.
Fig. 23. Propagation d. 10 temperature danl dOl porois a be.on lourd
de different.. .paineur•• a I'ulai ou feu normali.e apr', Y2, 1, 2
ef 3 heur••
-65-
",//// /'/f//.:////
//////
811d 25, Wood (JUS Sch luckenbatonvol lstemen nach einem Brandver.uch
von 3 Stunden Dever
Fig. 25. W(JII of solid slog concrete blocks after 0 fire 1est of 3 hours'
durotion
Fig. 25. Wall en blocs- pleins en beton de loirier de haul fourneou apr4U
un essni ali fe-v de 3 heures
Bild 26. Abnahme der S'andfe.tigkeit einer nicht eingespann.en Wand mit
der Bauhohe h
fig. 26. Reduction of the "ability of 0 nan-re.trained wall with ccnstrec-
tian depth h.
fig. 2'. Diminution de la re.istance d'une paroi non encadree de 10
hautewr h.
2
5
-- -- -- Paroi en b'ton leger de blocs creux, tipaineur 7,5 Clft,
hauleur 3 m (31)
Bild 27. Verwolbung von Wonden im Btand\lersueh
"'-- 15 em dicke Sehwerbetonw(Jnd 1,80 m hoch (33)
- -- - 7,S cm dicke Wand aus leichtbetonhohlblock..
• Ieinen Wandhohe 3 m (31)
Fig. 27. Warping of walls in the fire test
------ 15 cm thick dense concrete wall. 1.80 m high (33)
- - -- 7.5 cm thick wall of lightweight concrete hollow
block.; height of wall 3 m (31)
Fig. 27. Deformation de parais lars de I'euai au feu
Paroi en be.on lourd, epaiueur 15 cm, hauteur 1,80 m
(33)
1 2
Duration
3
! / I i
-1711
i / i
, ._--, · ~ - - ' I - - - + - - - - j
cm.....-_--r:----:---
-66-
lild 28. Tempcralurve,lauf in .inar S'ahlbolonplotlo wiihrond Nor-
mcnbrondes
Fig. 28. Tempera'ur. curve in Q rQinforcQd concrete slab during tho
Ilondard fir. test
fig. 28. Propagation de 10 t.",peratur. dans un. plaque en beton
"lIIt.n' I'tlua' • ., flu ".rmall.'
3 -hr-
Duration of fire test

I
j
I
E
E
S
:5
... ...... __ ' S?,
'"
1------ I
===: .... ,
l-. __

I
J) (?)
CJ) -:;
I
--- / /

/ 1/ -:
o / I -:
1//
V
v.Y


thOD
>::

OSOO
<+-<
>::400
-rl
(J)

§20
p'100
E
(J)
1:-4
from
500 o
Bild 29. Anstcigcn der Templ!rolurcn an den Bewehrungutiihlen von Stahl·
bctondcckcn chne und ",it Unterpulz
(1) uRvcrput11e Dcckcn
(2) 1,5 em Kolkzcmcntpul1
(J) 1,5 em Zement-Kalk-Vermiculiteputz
Fig. 29. Temperature rilG' in the reinforcing ban of r.inforced conere'.
f100n with and wilhout plasl.red underside.
11} unplasterod f100n
(2) 1.5 em cement-lime plasle,
(3) 1.S em cement-lime-vermiculite ploster
Fig. 29. Augmentation dc", lomporoturel QUlI armatur.. d. plofoRd.
belaR arm. sons et ovoc crop,nags
(l) Platands sans crcpiuoge
(2) Cropiuoge en eiment ef chou., epaiucur 1,5 em
(3) Crepillage en eimen., chou. 8' vermiculite, _paineu, 1,5 em
M/n.
110
(J)
I
G
90
i
-r1
"I
.j..)
i
(J)
0
>::
50
cd
.j..)
v:
-rl
[/J
3D Steel covering
(J)

Q.1lm
(J)

-rl
0
I'r.< 5 10 15 20 15cm
Slab thickness
8 JOO E- 225 OOOkgllm 2
I
Bild 30. Verandurung des innaren Hcbclarms :l infolgc ;(ndcrung de' Be·
tongulo unle, SORS' gloichcn VcrhCillnisscn (Schema)
Fig. 30. Variation of the inlcrnal lover orm 1: dUQ to change of the
concreto quality under ofhurwi5e equal conditions jdiogrommotic).
Fig. 30. (hoRgomant d" bras de lovier inferno z du au changement
de 10 quolile d" butan dans 10' m6mos condilians (Sdt'ma)
Bild 31. EinfluB der Oeckendicke auf die Feuerwidenlandldouer van Siahl-
betcnplouen
Fig. 31. EffClct of floor thickness on Ihe fire ,e,iuonee pe,iod of rein-
forced concrete ,lab,.
Fig. 31. Influonce de I'epaiueur d'un plafand en plaques de beloll
,,:ume sur 10 ,esislance au feu
-67-
time Test
10 20 JQ .0 50 60 70 80 901/itl.1Q/I
---I
70 80
time
50 60
Test
30 20 10
em
0
C
10
0 .'0
·rl
+.:>
30
o
100
())
H so
G-t
6C
())
c:::l
70
0
16L..---l_......._.....:.._...... _ ...... _....l.._....I.._....l..._...J
o
Jild 32. Ourchbiegung von Ferligleildecken im Normenbrondversuch.
Die lohl bez.ieht sith auf Tofel 4. Spalte 1
fjg. 32. Deflection of precast floors in the standard fir. test. The
ftt,llftber refers to Tobie 4, column 1
fig. 32. Deformation de pla'ands en beton manu'ach". a I·.uai au
"" normalise. L. chiffr. se rapport. ou tableou 4. colonne 1
Bild 33. EinfluB der Belastung auf die Durchbiegung einer Gasbeton..
deckenplatte im Normenbrandversuc.h nnch (37)
Fig. 33. Effect of loading an the deflection of on aerated concrete
floor slab in the standard fire test according to (37)
Fig. 33. Influence de la charge lur 10 deformation d'un plafo.nd en
plaques de betan.gaz. a I'essai au feu normalise selon (37)
in cm
:' """""
ection A-A
__

I
At-.-
Sectiol1A-A

dim.
I
lime
cern.
Section 8·8
Section '8-8
At---
I
Aild 34. Brcuduusb reitunq von eben ncch unten. Obeeseite r-iner Stohl-
betonhohldiele nc ch dem Brand (38}
Fig. 34. Downward spreod of fire. Top surface of reinforced concreto
hollow flooring unit after the fire (38)
fig_ 34. Propagation dl1 feu de- haul en bas, Cote superieur d'un
ale tend creux en betcn crme upres lessoi Ql1 feu (38)
Bild 35. Bctonfcrtigleile-Troppcn
Fig. 35. Precasl concrele stairs
Fig. 35. Escaliers en beron manufacture
-68-
I
V
I
• I
I Structural a
2... rno3."
rrime in mln. ,
!_--p.ar.t. ! j 20
&0 00 uo 1;0 1.:!
I
i - ICO -, I
I
"
c:==:- -'-:..-.J
,

fJJ--
I
J --- . .- .
j
. <:It
I 13
-,-------...,
i
,- "I
i--______._---....J
-20 _
I
i
I
I
-
'00 -,
J ,
, I
I
I
I
I
1,5 236
--
-_._-. -- ,- _.-
J I'
Ei5I I '3
1--.------
1
I
_}O r- DI1'h inC"'"
I·'
&ild 37. Einflull dolo Allors yon Stohlbetonbalken auf dio Feuerwidorllond.,
douor
Fig. 31. Effccl of tho ogo of reinforced concrete beams upon the f;,.
resistance pcricd
Fig. 31. Influonce de 1'6go de pcvtrcs en bolon orme sur 10
au fou
2 J hrs. 4-
resistance time
j,
5
o
'H
o
s,
QJ
:>
o
u
bD
c
·rl
o
s,
o
'H
c:
·rl
QJrl
'-' J
QJ.p
,.Clf)
.p ;,:
Bild 36. Einllufl der lf bcrdc ckvnq der Stahloinlogon auf die Fouor-W.dor..
sland,douer Yon Stohlbclonbolkcn
o 0 britiHhc Vcrsuchc
nic dcrlc ndis che Vcnucho
• • dculschc Vcnwchc
Fig. 36. Effect of the cover of tho roinforcing steel upon .he firo
resillonco period of reinforced concrele beams.
o 0 British tcs ts
Dutch tesls
.----. German tcSls
Fig. 36. Influence do 10 couvorfurc sur I'ormature en Geier do poulre,
en beion ormc 10 resislonce au feu
o 0 Euais britanniques
X X Essais neerlandai.
• • Euoi s allemands
Bdd 38. BrancJvcnuchc an Stohlbctcn-Plottcnbclkcn aus 'Icrschiedoncn:
schlogsloffcn (28). Bolkcnqucrschniu
Fig. 38. Fire tests on reinforced concrete T-beams mode with veri;.
aggregates (2a). Cros s-scetien of beom
Fig. 38. Essc! au feu sur dolo povtros en plaques en belan arma c.
diyors agrogats (2S) section des powtres
I
t
0
l 1
,
f-l0 /
I
_.v.:> ('"
"1.
M
lild 39. Fcucrwidcntondsdauer yon rechteckigen Stohlbetonsawlon in Ab-
hongigkeit vern Ouc rs chnitt und Zuschlogs,offen
!\ 0 quarzitischer Zuschlog
A • Kolkstein-Zuschlag
f,g.39. Fire rosivtcnce poriod of rectangulor roinforced concrete columns
0\ 0 function of cron·section ond oggregOles
.. a 'quorfzitic a99r;;90to
" • limestone aggregate
39. Resistonce ou feu do cetcnnos en belon arma rectonguloires on
lonelien de 10 section et des agrogals
. a Agrcgat cvec quarlxite
• Agrcgal avec colcaire
Deuhcho Venucho; Oroiccko: Britische Vorsucho
4 hr-s ,
ime
J z
c
E 200
::J
rl
o
U
em'
§1000....--------,----------,
·rl
.p
O 6JO -
Q)
Ul
I
Ul 500 .--
Ul
o
t; 'tOO
Dim.in cm,
I
"1 r- . '6,5
-69-
Sorie 611.7
r
Loo p
6:mn
e=8cm

S.,i. J", 4
1"'-,- 20
'r
1
o
I
r-------.!-ooP
r ,...------_
, I! I'
".,."' !. e. =8cm
'" I " ... I,
i ; / . <he sh
I <------".t
i r e
i
Steel covering
of loops,...2.5 cm
Bild 40. Querschnilte der Fertigteihtutzen fur die Brandversuche des
desverbandes der Betan5teinindustrie
Fig. 40. Crass..sections through precast columns. fOr the fire tests of
th<l Bunde5verband der Betonsteininduslrie (German Precast Concrete
federation)
fig. 40. Seelion des appuis prefobriques destines cux essais au feu de
I'onociotion des fobriconts de betan manufacture.
Bild 41. Stchlbetonstiitxe au. Kolk.teinbclon mit Ma.chinendrahleinlage
nach dem Brandversuch und ans-t:hlieBender loschwasserprobe
(Ausschn.)
fig. 41. Reinforced concrete column made of limestone concrete with
incorporated wire mesh cffer the fire fesf foltowed by sprayit19 wifh
wafer (parf view)
fig. 41. Appui en beton orma. avec colcoire et filet en ocier comma
armature epre s I'essoi ou feu ef ensuite crrcse avec de I'eeu.
Bild 42.EinfluB von Betonquerschnitt und Bewehrungsverholtni. auf die
Feuerwiderstondsdouer von Stahlbetonsoulen
deutsche Versuche; quan.itischer Zu.chlogsloU
= 2-3010, Soulenlange 3,60 m
- -- - deutsche Versuche; Kolkstein
= 2_3°/0, Soulenlange 3,30 m
-x-x-X-X ftanxosische Versuche; quarzitischer Zuschlog·
stoff = 0,6--6,8°/0, Soulenlongo 2,30 m
Fig. 42. Effcct of concreto section and reinforcement proportion upon
the firc rosistance period of rcinforccd concrete columns
German tCs.u: quartzitic aggregate
jJ. = 2_3°/0 j length of column 3.60 m .
-- -- -- German tests; limeslone
jJ. = 2_3°/0; length of column 3.30 m
__x __x__ French lests: quartzitic aggregote
- 0.6--6.8°/.; length of column 2.30 m
Fig. 42. Influence de 10 section du bdton et de I'armature de colonnes
en beton arme sur 10 resistance au feu.
Essais allemands, agregat avec quartJ:ite
= 2_3
J
/o, Longueur de 10 colonne 3,60 m
- -- -- Essais allemands; Calcoire
""" 2-3°/0; Longueur d ela colonne 3,30 m
-X-X-X-X Essais frant;ais, agregot avec quartzite
= 0,6--6,8°/.
r
; Longueur de 10 colonne 2,30 m
+ hr s
time
2 3
resistance
o

o
o
>::.",
3IOGOr----,---.....,..-----::r:-----,
rl
o
U BOO 1-----\---- 1--
G--i
o
600I----\-----F--r--t-----j
>:::
'0
.,.-j.co
+..:>
o
OJ
tr: 200 t-----j-
-70-
Bild 44. Ein'luA von Bolkonquorlchnitt und Stahluberdodtung auf die Pecer-
widerllondldauor von SpannbofonboUren mit lo'orl;gom Verbund.
(God rung one und Quoruhnillo)
Fig. 44. Effoct of beam lection and COYor to Itoel upon tho 'iro
rOliltonco period 0' proltrolled beam. with pre-tenlioned tondon••
(Squat lectionl and 1.lectionl)
Fig. 44. Inlluence de 10 loction ef do la COUYorture lur .'armolure de
poulrel en b'Ion pr'conlrainl, oliomblOo. immediatement, lur 10 r'li-
Iionce au 'eu l'e'ite. lection. et en 'orme I)
fSOOem
l
section
1000
beam
500
Hin
f5
0
r-- - - - ......,- - - - - ......,- - - - - --,
steel covering
-ri
flO
_ a 1 -1.5cm
liililil 0 5 em ,
QJ
090

m

-ri
UJ
QJ./O

O'-------:-:':,...-----...l---.:.---J
·ri
rr,
-8 00 J6 10
Bild 43. Brandvenucho on unbowehrten Boton,tullo" nam (44)
Fig. 43. Fire tel" on plain (unrein'orced) (oncre'e columns according
'0 (44)
Fig. 43 Ello, au feu 6 de' oppuil en beton non arme ,elan (441
ill
..........
S
/' ......
" , -,-l
/ \
.j...)
/
\
QJ
I
\
0 '.

)
enu of test ciue
m
.j...)
to excessive heat
UJ
-ri
transfer
UJ
01

QJ

-ri
0 p'
'bOY. c.-;
end restraint
fOf)()
800

...
600

.,
...

'"
200
0
: bur'rr-ou t test
heating
o C'
5 6
hours
Bild 45. Wirkung eincr Endciospannung auf dio Fcucrwidcrstondsdaucr
von Spcnnbetcn [schcmcfis ch] [51)
F;g. 45. Effect of end restraint upon tho fire rC5islol'\cC period of
prus rros sed concrete (diagrammatic) (51}
Fig. 45. Effot d'vnc fixolion d'un betcn precontroint 10 rthistoncc
au feu (Schomotique) (51).
Bild 46. Temperotuneilkurven 'ur AUlbronn- und Heil.venuche on Schorn-
Itoinen nach DIN E 18 160
Fig. 46. Time-tomperature curve, 'or burn-auI lelll and healing lelt.
on chimnoy. according 10 DIN E 18 160
Fig. 46. Courbel de tomperature-Iempl pour del e..ail d"chauffemonl el
ClU 'eu aur: cheminoel lelon DIN 18160.
-71-
./
Bild 48. Wand GUS Gtosbousteinen wahrcnd des Normenbrandversuches
Fig, 48, Wall of qloss building block. during the .hmdard fire le,1
Fig, 4ft Paroi en b locs en verre durant I'euai au feu normalise,
0
/', :>....
~ "j' /
/.."/"
I I ! I
L
Bild 47. Bcis picle vc n Schcr nsteinfcrmsnickon c u ~ Loichtbctcn
Fig. H. Examples of lightweight concrete chimney units
Fig. 47. Exemple s de pi e ce s en beton leger pour cheminees,
-72-
Bild 49. Verkleidung cus Asbeslzemcnlplalten schulzle gegen Flugfcuer
Fig. 49. A facing of asbestos cement slabs afforded protection against
flying sparks
Fig. 49. Porcmenf en plaques demicnte-ciment protegealr centre un feu
emporte.
Bild 50. Stllhlbetonbalken mit unlerschiedlichem Bugelab'lllnd naeh einem
Schadensleuer (nach 58).
Fig. 50. Reinforced concrete beams with varying stirrup spacing otter
a lire (Irom 58)
Fig. 50. Poutres en b"lon arm" a <Iillerente, distances de I'etrier apr••
un incendie (.elon 58).

NATIONAL RESEARCH COUNCIL OF CANADA Technical Translation 1317

Title:

The behaviour of concrete products and precast reinforced concrete members in a fire (Das Verhalten von Betonwaren und Stahlbetonfertigteilen im Feuer) H.J. Wierig Betonstein-Zeitung, 29 (8): 395-~07; (9): ~~3-~5l; (10): 503-510, 1963 (Reprinted in Untersuchungen an Beton, 6: 3-31, 1964) D. A. Sinclair, Translations Section, National Science Library

Author: Reference:

Translator:

THE BEHAVIOUR OF CONCRETE PRODUCTS AND PRECAST REINFORCED CONCRETE MEMBERS IN A FIRE

1. Introduction Fire is one of the natural forces to which man has always been exposed. From antiquity down to the present, history tells of many catastrophic fires which destroyed whole villages and cities. At an early date attempts were being made to profit by the experience gained from fire and to adapt the construction of buildings and to plan villages and cities accordingly. As an example we may cite the well-known fire of London in 1666. At the time of reconstruction of the city, very extensive building regulations were enacted(l). For example, the kinds of building materials to be used were specified, and the width of the new streets was determined. These regulations can be regarded as the forerunners of our present-day regulations. However, systematic research into the behaviour of building materials and structures exposed to fire was only begun about 100 years ago. condemned to demolition(1,2). At first the investigations were restricted to the burning down of old buildings Later, systematic fire tests under wellToday, in most countries, there are defined conditions were introduced. is under preparation.

well-defined standards and test procedures, and an international standard In Germany fire tests are carried out under DIN 4102 This A draft of the testing regulations "Resistance of building materials and members to fire and heat". standard is at present under revision.

contained in the revised version of this standard has just been pUblished. In the last 100 years a great deal of knowledge has been gained about the behaviour of building materials and members in fires. Nevertheless, In the damage caused by fire has continued to increase even in recent years.

Figure 1 the finely hatched columns show the increase in damage due to fire in the Federal Republic of Germany during the years 1954 to 1961(3). The growth of the gross national products reduced 1000 times is shown for the same period by the coarsely hatched cOlumns(4). product. It will be recognized that the Figure 1 In this damage has been increasing at a somewhat faster rate than the gross national Similar trends have been noted in other countries. illustrates very clearly that all economically feasible measures must be taken in order to prevent any further increase of damage due to fire. tance. Structural measures include the following: connection preventive structural fire protection acquires a special impor-

e.e. time can be attained in a given member. quite different conditions prevail. i. is only of secondary considFor building compononents which have to withstand comparatively This is the case. suffer an alteration of important mechanical properties as an effect of exposure to elevated temper- . 2. i. At higher temperatures. burn and are destroyed. briefly. Three different points of view must be high temperatures while performing their design functions. The state of the structure after only a simple safety measure is required. for example. including concrete and steel. and provision of escape routes for the use of the inhabitants and occupants of a building in the event of a catastrophic fire. two basic cases must be distinguished. Wood and many plastics.) eration. of the outbreak of fire~ measures hindering the spread of fire. and the posqibility of improving the fire resistance time of individual members. and more recently in reactor design. they begin to Not so well-known is the fact that the "non- combustible" materials. In considering the behaviour of structural members in fire. Designing architects and engineers must have sufficient know- ledge of the construction possibilities by which an adequate fire resistance. for example. structural design. but very severely exposed. different regulations are applied to waresin~le-family houses and theatres in the building codes than to small. For example. they must continue to exercise their intended functions. as far as possible. In many respects there is no fundamental difference between the behaviour in fire of precast and poured-in-place concrete construction. precast reinforced concrete members.-4the prevention. 2. The required fire resistance time depends on the nature and function of the building involved. For the prevention of the spread of fire damage the load-bearing and separating members of a building must resist the fire for a sufficiently long time. During a fire a construction is only (For this unrepeated catastrophic case. with a view to continued use.1 The Behaviour of Building Materials Exposed to Elevated Temperatures General With respect to the behaviour of building materials exposed to elevated temperatures. in furnace construction. the fire. fire and will discuss from individual examples the special properties of concrete products. are amon~ taken into account both in the selection of the building materials and in the The present paper reviews the behaviour of concrete and reinforced concrete in the "combustible" building materials. dwellings. combustible and non-combustible materials. we can separate two broad classes.

. We shall now review the most important changes that occur in concrete and building materials which are used in conjunction with concrete. 4. elongation at fracture and creep behaviour. While at room temperature the mechanical properties of the solid concrete are determined primarily by the properties of the cement paste*. The texture of a specimen of mortar and concrete can remain intact in the presence of increasing temperatures only if the resulting changes of shape of the separate components are mutually compatible. Regarding the behaviour of structures in fire. is greatly influenced by the aggregates as well. behave best. 2.-5atures. and the internal stresses can be absorbed by the cement paste. the following IJrOperties of the building materials and their changes due to higher temperatures are of importance: 1. smooth as possible. The high temperature behaviour can differ considerably.? 2.21 Mortar and concrete Thermal expansion Mortar and concrete are made up of cement. ~n 3. strength. modulus of elasticity. Coefficient of thermal expansion. many rocks crumble on exposure to extreme heat. 2. yield point. thermal conductivity and specific heat. of the aggregates. Moreover. For example. the high temperature behaviour. Hence. depending on the mix. due to the effect of higher temperatures. let us first consider the behaviour * hydrated Port land cement paste. aggregates with a low coefficient of thermal expansion and a temperature-expansion curve that is as Figure 2 shows the thermal expansion of Therefore. since they make up the greater part of the volume of the concrete. The effect of the aggregates on the behaviour of mortar and concrete at higher temperatures depends primarily on their coefficient of thermal expansion. the materials employed and especially the aggregates. aggregates and water. especially the coefficient of thermal expansion. changes in composition and structure of some building materials can affect the behaviour of the construction a fire.

Table I contrasts the various test conditions. then. according to ref.-6various aggregates as a function of the temperature(S. basalt and a number of kinds of limestone are found to have favourable coefficients of thermal expansion. Structural Of course. however. Contrary to experiments with small specimens. The numerical data of these It is clear. increasing cement content(7). compositions of the cement paste (water-cement ratios) and different rates of heating and different sizes of specimens. that first a thermal expansion A further reduction of volume takes place on Probably the different results are due to different Even for basalt. If the cement paste is heated not once. Figure 3 shows the change of volume of cement paste from Portland cement according to Endell(S). but several times. Among natural rocks. When mortars or concrete specimens are heated it is not only the aggregates that expand. the thermal expansion occurring in concrete is less than the aggregate itself. which under laboratory conditions takes place at S73°C with a heat consumption of approximately 8 cal/g. giving off gas. structures in the fire are only slowly heated from the outside in to a temperature above SOOoC. Since the aggregates expand when exposed to heat. which is subsequently compensated for and exceeded by the shrinkage due to release of water. Under otherwise similar conditions the coefficient of thermal expansion decreases with In Figure S the thermal expansions of various . cooling after heating. if concrete constructions that have been exposed to fire are to be used again.(6) . the practical effects of quartz conversion is not as serious a matter in a fire as it may appear to be at first glance. expand markedly. a reversible positive expansion occurs from the second heating on (Figure 4). expansion curve is due to the a~S It will be noticed that for quartzitic rocks there is a characteristic discontinuity of the This bend in the temperaturequartz conversion. The cement paste also changes in volume. To be sure. non-quartzitic rocks generally behave better than quartzitic material. at approximately 900°C. investigators differ. thermal expansion between SOO and 600°C. while artificial stones similarly endowed include blast furnace slag fragments and crushed brick. and Philleo(7). At this temperature the rock begins to occurs. relaxation. With the exception of diabase. proceeds only to a limited depth.6). therefore. there is a critica~ temperature. the affected layers must be r-emov ed . whereas the cement paste begins to shrink again at higher temperatures. Nekrassow(6).

The scatter range For a special for the tensile strength is horizontally hatched in Figure 6. British investigations(16) showed that after exposure to temperatures up to about 300°C for several weeks no further decrease of strength could be expected (Figure 7). which presumably can be attributed to the above-mentioned severe internal stresses arising between cement pastes and aggregates. of the same kind do not necessarily behave the same. The initial increase of strength on heating of certain concretes can be explained by the fact that heating brings about a "steam hardening". All stones Considerable variations are possible among quartzitic material. concrete tensile strengths drop more rapidly than the compressive strength values. a reaction between cement and aggregate is by no means impossible. time of testing after heating. etc).lS). storage conditions.12. age of the specimens. heating times. water-cement ratio. represented in Figure 6 by the . however. One group already found a decrease in compressive strength for the comparatively slightly heated stage of about 200°C. are interesting.22 Strength When specimens made from mortar and concrete are exposed to higher temperatures their strength values are altered. according to which the waterOf greater influence. fireproof concrete. Other tests. however. The investigators fall into two groups. The increase of strength in the temperature above lOOOoC. ratio of cement to aggregates. showed an increase of compressive strength up to about 300°C and a drop only on further heating. and especially among limestones. 2. The data of Malhotra(ll). dictory. which continued at a more or less steady rate at higher temperatures. type of aggregate. mix. leading to additional hydration of the cement. With many types of aggregates. and secondly to differences in the initial material. According to French investigations(17). is the cement ratio has no appreciable effect on the relative change of thermal compressive strength. the compressive strength may be greater after very severe heating than those indicated in Figure 6(6.-7mortars are plotted as functions of temperature after Endell(S). moreover. Publications(6. construction of test furnaces.8-1S) on strength tests of mortar and concrete in a heated state are partly contraThis is due on the one hand to varying test conditions (sDecimen) size. The vertically hatched area in Figure 6 shows the scatter range within which most of the results of concrete compressive strength tests at higher temperatures fall.

19) a slower rate of decrease of the modulus of elasticity was found l)y Cruz and Philleo. If we con- . Woolson 1s(18) Figure 8 shows results on diabase and limestone concrete.24 Heat conduction and specific heat The internal heating of a structural member depends largely on the heat conductivity and the specific heat of the building materials employed. and Busch. This strongly affects the deformations in a Also affected is the interplay of external and internal forces in statically indeterminate structures. For practical application. also. depending on the nature of the stress.12. This means that the ratio of E co Id/E warm is more favourable when the tangent moduli are compared. in these tests. 2. in prestressed concrete constructions and even in standard. 2. dealt with cube strengths of 280 kg/cm to 50 kg/cm 2 • concretes with a cube strength of about 160 kg/cm 2 2 • Busch's investigation The compressive stresses were 30 According to Woolson The very strong decrease of the modulus of elasticity even at 0 relatively low temperatures up to 400 0 c is striking. later confirmed Woolson investigated • by Busch(9) on sandstone concrete (quartzitic). investigations are also plotted in Figure 8. the modulus of elasticity at 400 room temperature. is due to the fact that in properly composed concrete the lost bond of hydration is replaced by a ceramic bond. therefore. again fell to 40 to 50% of its value at room temperature. reinforced concrete cross-sections. but depend on various factors. a decrease in the modulus of elasticity occurs. structure. not. The difference between the results of Woolson and Busch on the one hand and those of Cruz on the other may be sought in the fact that according to Busch's findings the modulus of elasticity of a concrete that has been exposed to fire increases with increasing load. the secant moduli. c is only about 20% its value at In more recent investigations(7.-8broken-line curve. These are not constant values. the modulus of elasticity at 300 oC.23 Mortulus of elasticity When concrete is exposed to elevated temperatures. values may be chosen which lie between those of Busch and Cruz. The results of these However. depending on the water-cement ratio. Some existing results of modulus of elasticity measurements at higher temperatures differ sharply from each other. as usual.

Figure 9 shows these effects at room temperature(20). we must distinguish between three basic temperature ranges: Range I Range II Range III be10w oOe between 0° and 1000e above 1000e.-9sider the heat conduction of concrete. however. The practical importance of the increase of thermal conductivity of concrete with the temperature is limited.25 kcal/ kgOe. In range II the heat conduction (kcal/mhOe) of the concrete is influenced primarily by its bulk density and its moisture content. Normally. therefore. the initial temperature of a construction at the start of For the sake of completeness. In Figure 10. 6T The heating of a thick plate subject on one side to temperatures which vary in time is governed by This is written 6 2T 6s 2 IT 2 = a . it is shown that the thermal conductivity at constant bulk density depends not only on the moisture content.21 . Here again there is a definite rise in the thermal conductivity with temperature.23). The specific heat of regular air-dry heavy concrete at room temperature can be taken as 0. is less than in range II. but with partial filling of the pores. In temperature range III. of aerated concrete is attained not for total saturation with water. Fourier's equation of heat motion. mentioned that a considerable delay in the internal heating of the structural members is due to the latent heat of melting of ice. In conjunction with the behaviour of concrete structures in the presence of fire. concretes are completely dry. . it must be a fire will be above oOe.0. may also be inferred from Figure 10 that the maximum heat conductivity value In Figure 11 the thermal conductivities of a few concretes at higher temperatures are represented as a function of the temperature(22. range I is of interest only in a very few exceptional cases. tics increase. The coefficient of thermal conductivity. with the example of an aerated concrete having a bulk density R = 520 kg/m 3 the temperature(21). but also on At higher temperatures the heat conduction characterisIt This is due to the fact that at higher temperatures the amount of heat transferred by radiation in the pores increases greatly.

ration of water. thus The spontaneous liberation of CO 2 by the limestone sets in at approximately 900°C. 3 The latent heat of evapoThis is a very large amount It follows that the behaviour of a concrete construction in a fire. the thermal diffusivity which depends on the quotient }.-10where T t a 2 temperature (OC) time (h) coefficient of thermal diffusion (~ ). The vigorous liberation of water Above 400°C a slight loss of weight then occurs due to liberation of water from the calcium hydroxide of At about 600°C the conversion of the limestone begins. For heat transfer.12). the cement paste. This may be summed up as follows: The thermal conductivity of concrete is determined primarily by the bulk density and the moisture content. is decisively influenced by the water Here quantities 2 into CaO and CO at a rate of nearly 400 kcal/kg CaC0 3 at a surface exposed to the fire. The greater the bulk density and The complete heating of moisture content. therefore. Figure 12 shows the loss of weight of a concrete containing some calcareous aggregate according to ref. content. between 100 and 300°C is clearly evident. of course. is over 500 kcal/kg. of heat are also consumed in the conversion of CaC0 delaying the heating of the structure. A parallel phenomenon occurs in limestone concretes. dC Y bulk density (kg/m 3 ) Since the specific heat c of the concrete also increases within increasing temperature(9.(7). 2 s = depth in the wall (m) 2 (thermal diffusivity) . and especially its internal heating. but at low CO 2 partial pressures it may start as low as 600°C. Much more decisive is the delay in internal heating in a fire owing to the evaporation of the water in the concrete. the greater the conductivity. but the thermal diffusivity a follows: where a 2 which is defined as = coefficient of heat diffusion (~ ) 2 (thermal diffusivity) thermal conductivity (: kcal °c ) c = specific heat (kCal )m· hkg . the decisive factor is not the thermal conductivity. . of heat. varies but little.

The values of Figure 14 are important for the behaviour of the material during a fire.4% carbon is plotted after ref. It will be recognized that the thermal expansion The coefficient of thermal Compared with the thermal expansion of the Generally is continuous up to temperatures of about 700°C. 2.(24). The change in the mechanical properties In the following paragraphs a number of of steel on exposure to high temperatures depends on the kind of steel and typical examples are cited inasmuch as these are required for an understanding of the behaviour of reinforced concrete constructions. The hot tensile strength curves have been left out.32 Yield point at elevated temperatures Figures 14 and 15 show the yield point (0. concretes (Figure 5) the expansion of steel is somewhat less than sand-andgravel concrete. same even over considerable temperature ranges. Rolled steels in general recover their original strength on cooling. but somewhat greater than limestone concrete. Figures 14 and 15 differ in that in the former the tests were carried out in the hot state.-11- structural members is affected decisively by the evaporation of the free and bound water. those of Figure 15 for appraising the bearing capacity of a structure after a fire. 2. It is clear from Figure 14 that the yield point decreases at first speaking.31 Thermal expansion In Figure 13 the linear thermal expansion of a steel with 0.2 proof stress) for a number of steels as a function of the temperature. including steel concrete and prestressed concrete.3 Steel The fire resistance of all reinforced concrete constructions. 2. however. because by and large they are closely related to the yield point curves. expansion increases slightly. whereas cold-forged steels lose some of their strength. especially on the carbon content. is decisively affected by the behaviour of the reinforcement. the thermal expansion of the steel and the concrete is the . Only in the temperature range around 200°C is there at first a slight rise in the tensile strength. while in Figure 15 they took place after cooling.

34 Modulus of elasticity Figure 17 shows the change in the modulus of elasticity of a nonsiliceous S. Contrary to a widespread misconception. after ref. mild steel as a function of the temperature. Figure 8 shows the heat transfer It must coefficient of pure iron and of structural steel with 0. M. 2. be realized that at comparatively low temperatures. The behaviour It is clear from Figure 17 that the modulus (27).e. especially where pre- . at the beginning of a fire.33 Rupture strain Figure 16 shows the change in rupture strain of a mild steel under the influence of elevated temperatures.-12slightly with increasing temperature. o At 500 to 600 C the design stresses calculated in the construction are attained. 2.8% carbon content. therefore.oC 0. 2. The values for most structural steels lie within the hatched area. of other steels is similar. the yield point of high quality prestressed rods is not below that of ordinary reinforcement rods. about twice as great as at 20°C.4 At room temperature the specific heat is approximately c kcal kg. the heat transfer coefficients of steel are 30 to 50 times as great as that of concrete. It is only in the temperature range of around 600°C that all the values more or less coincide. The decrease. temperatures it increases just as rapidly.35 Thermal Conductivity and specific heat The thermal conductivity of the steels normally used in construction decreases with increasing temperature. i. Up to 200°C the rupture strain at first decreases rapidly. This fact affects the behaviour of very heavily reinforced structural members in the fire. but then in the region of higher At 600°C the failure strain is of the steel drops to about half its value at room temperature between 500 and 600°C.11 and at 800°C it is 0.2 Other materials In the erection of concrete structures.uc 2. is not as great as in concrete. The specific heat of iron and steel increases with increasing temperature. but then more rapidly. kcal kg .

owing to their poor contact with the atmospheric oxygen. Ceiling fillers or slabs of mineralized wood wool have been found excellent in conjunction with concrete. No general Under unfavourable conditions. thin Caution is always required whenever parts of the whole construction which are essential for fire protection are secured to wooden A critical temperature for the ignition of wood can only be stated approximately. even lower ignition temperatures have been at 250 to 300°C. development is in full flood.g. . avoided. plasters or for sealing purposes. or the like. sections always have a higher resistance to fire than delicate. since even in a fully raging fire they ash slowly. in the form of small structural parts such as dowels. with gypsum plasters under very gas-tight floors the steam pressure may become so great that the entire plaster breaks away from the ceiling. however. Gypsum.g. the surface texture and other factors. and plastics. ignition may be expected Some times. to securing projecting sections. These materials include. its moisture content. when rapid heating of a building element must be However. considerable Use can be made temperatures. When gypsum is exposed to higher For this. e. parts. as is known. Moreover. veneers. for example. In what follows we shall briefly outline the most important prop0rties of such materials with respect to fire. since it varies greatly with the kind of wood. The resistance of thin steel reinforced concrete ribbed ceilings to fire can be greatly improved by the use of these parts. quantities of heat are consumed. hardens by hydration of the semihydrate or of the anhydrate to the dihydrate form. structural parts. Wooden laths are also embedded in the concrete in the manufacture of certain concrete products with a view Wooden parts surrounded by concrete on In addition. after dehydration the gypsum loses much of its strength. In the field of plastics. the atmospheric oxygen available. a number of other materials besides concrete and steel may be employed in the finished structure.~1). some water is given off between 100 and 160°C and complete dehydration to anhydrate occurs at about 200°C. e.-13fabricated concrete parts are used. gypsum products. wood or wooden wares. as bonding a~ents. of this fact at times. larger crossseveral sides do not ignite as easily as completely exposed ones. noted. Wood occurs in concrete structures. and thus constitute good insulation(29. Hence the heating of all constructions made of gypsum is greatly delayed at approximately 100°C.

where plastics are popular as binding agents. 3. In many countries there are standards for conducting fire tests. which fits However. unless a temperature equilibrium sets in before hand. considerable deformations or stresses occur. because subsequent changes are frequently very difficult to carry out. 3. therefore. which. a fully developed conflagration. The rapid changes of temperature at the beginning of a fire in At As general result in large temperature differences in the structural member. differences in the results from fire tests in different countries have been due less to different temperature loads than to differences in test material and other test conditions. even under earlier conditions. Basically. fireproof requirements must be taken into account in the planning stage. adhesives or joint seals.-14critical temperature for all plastics can be stated. but more even heating of the construction.e. Figure 19 It will be noted that all the curves are very similar to each other. the previous curves well. Some plastics alter Especially in the field of prefabrication. so that the results on constructions investigated by the standards of various countries on the whole are comparable with respect to temperature load. In attempting to analyse the curves represented in Figure 19 it must first be realized that in the first half hour. shows the temperature-time curves used in various countries. i. there is a continuously progressive. a uniform curve has been drawn by ISO. and then increase more slowly. the temperatures in the fire compartment rise very rapidly. therefore. it is laid down that the structural part to be investigated must be exposed to damaging fire(3 0). although. In order to get the best possible agreement. at the start of the fire. These regulations are alike in principle. they differ from each other in detail. In order to get comparable test results it is customary to represent the development of an idealized fire by temperature-time curves.1 The Behaviour of Concrete and Steel-Reinforced Concrete Members in Fires General A knowledge of the change of properties of building materials will The suggest important ideas concerning the behaviour of structures in fire. can result in failure. their properties even at lower temperatures than wood. however. fire test on the whole structural part. an early stage. The uniform temperature-time curves correspond approximately to the course followed by a fire of medium severity. is decisive in appraising fire resistive quality. . the fire continues.

which may by chance be stored on the unexposed side of the space separating members. Essentially. such conditions would occur general. or a time test may be made. such as beams. must retain their bearing strength and stability. fire test. are destroyed when the temperatures go higher than this. i. The stability is threatened by severe deformations. because for the most part the temperatures lie below the 1200 0C limit. no hot gases should seep through. it is determined whether the functions will continue to be performed during a predetermined time. they should not afford a passage for the fire. however. concretes. only in rare and exceptional cases. In the course of a fire the structural parts ought to perform their functions. Space separating structural members must not allow the fire to pass. kinds of natural stones. There is little In difference in the regulations on this point from country to country. columns. is regarded as adequate as far as bearing capacity is concerned. the simple calculated permissible load is applied in the Since fire is a catastrophic occurrence. in the case of walls exposed to the fire. and at individual points up to 180°C is allowed. which show excellent behaviour in fire up to about 1200 0C owing to their good thermal insulation properties. for otherwise the fire will . etc. the structures may be tested to failure.e. the resulting deflection may produce such great eccentricity that the wall will collapse.e. on one side. these are as follows: a) Bearing parts. as the case may be. and the temperatures on the side away from the fire must not rise above a certain value. decomposition or deSeveral Many light-weight struction of a number of mineral building materials occurs. In order to be able to estimate whether and how long the bearing capacity of a structure will remain intact in a fire. however. two cases must be considered in the testing of structural members for their resistance to fire. until they no longer satisfy the above functions a or b. The reason for this requirement is to prevent easily ignitable material. for example. i. i. b) Space separating parts must continue to be effective in separating space. the resistance of bearing structural parts to fire is tested under load. from igniting. On the one hand. At this temperature. That is to say.-15This is important.e. ceilings. The critical rise in temperature in most countries today is put at an average of about 140°C above the initial temperature. Of course. a safety factor of one For example. decompose. Basically.

The duration of the fire test depends on the type of structure in which the test member is to be used. heat equilibrium is more or less attained.g. we must distin~uish blocks. To carry out the fire test. still longer resistance times are prescribed. between bearing and non-bearing . or walls of glass blocks. buildings such as warehouses. which are particularly exposed to fire hazards. various countries. Increasing the demands for duration of resistance against fire beyond a certain time does not always require substantial changes in the construction. doors. according to DIN 4102 must continue In other countries. 3. e. a fire test chamber for ceiling constructions.21 Walls of light weight and heavy concrete Heating Concrete walls can be manufactured either monolithically from heavy or li~ht weight concrete. Special demands are made of chimneys and roofing materials. "fire- resistant". or may consist of slabs of considerable size or of Basically. Requirements differ greatly in the In Germany for buildings threatened with fire. and. for walls. This is because for many structural members there is a In this state the heat losses on the side away certain time at Which. is reqIJired. semiToday the schematically. The wall is prepared for testing. In a number of countries some relaxation of these rules is allowed for certain members. the structural member to be tested is installed in a fire test chamber. from the fire are approximately equal to the input of heat on the side If the structure has not collapsed by this time its resistance of fire can be greatly increased by reaching this threshold.2 3. for certain to perform their functions in the fire tests for 1 1/2 hours. if it is a bearing to a load. testing institutions of many countries have very modern fire test chambers The test member is installed in the fire test chamber in a manner that simulates as closely as possible the incorporation of the member in practice. towards the fire. during the standardized fire test.-16jump the barrier. membe~ is subjected Figure 20 shows a fire test chamber with a loading apparatus Figure 21 shows. and "highly fire-resistant" must do so for 3 hours. parapets. with partially or fully automatic control. and in exceptional cases "highly fire-resistant" construction "Fire-resistant" members.

time F increases exponentially with the wall thickness d. the joints have to be profiled as far as possible. due to thermal insulation considerations are not endangered by fire. only slowly go above lOOoe. of course are liberated again during condensation. generally speaking. so that adjoining plates fit together in tongue and groove fashion. approximately according to the function F = d l. so as to ensure that no hot gases can penetrate through any gaps Especially where thin plates are employed. so that no increase of temperature ~reater than 140 0c above the initial temperature will occur on the unexposed side. in the case of heavy concretes and wall thicknesses of d < 15 cm. careful design and execution of the joints is extremely important. considerable Thus quantities of heat are needed for the evaporation of the excess water.-17walls. the The fire resistance deciding factor is the thickness of solid material. In the case of walls with cavities. can happen that the shells on the side of the fire will burst after a certain time and that cracks will form through the entire block as the test continues. in unfavourable cases it the temperature rise are observed on the side away from the fire generally for temperatures between 80 0 and 1000e. even in the case of thin walls. the kind required for outside walls in tall building constructions. Grouting or smoothing of the joints may help. whereIn general. First of all. in the joints. 7 ( 31). For masonry brick walls or slab walls. In the case of thin walls. however. which quantities. The thermal insulation of the structural members during a fire is greatly affected by the moisture content. there is a possibility that the permissible rise in temperature on the side away from the fire may be exceeded. but In the case of concrete separating members. Figure 22 shows the necessary thickness for massive walls of heavy and light weight concrete as a function of the test time. A few kinds of As hollow blocks made of heavy concrete constitute an exception to this. i~ in the case of hollow blocks of other materials. . Moisture often penetrates in the form of water of condensation The temperature of these moist In the case during a fire at joints and hairline cracks. structural members with high moisture content rise rapidly to lOOoe. thick walls of as the latter operate only as space dividers. such as walls and ceilings. The former have both bearing and space separating functions. lags places is somewhat higher than that of their dry surroundings. The critical increase of temperature on the side away from the fire up to test times of 3 hours is only important. The permissible rise in temperature for light-weight concretes is seldom exceeded.

other. Numerous compilations about fire tests on walls are available in the literature. Figure 25 shows a wall of slag concrete blocks after a 3-hour fire test.temperature on the side towards the fire 1 - 8d T . since the scatters and temperature measurements inside the walls are very great owing to differences in moisture content. the bulging depends on the material. The two diagrams are indicative only. because fire tests on walls are usually made only on spec i- . Table II contains a selection of a few typical walls.22 Deformations and stability When exposed to fire on one side. walls tend to bulge more or less extensively towards the fire side. the wall thickness. The delay in heating at 100°C is definite in the case of the aerated concrete wall.height of wall T .-18of a heavy concrete wall with a bulk density of about 2400 kg/m 3 . is four times as great. and the bulge may even appear on the other side.T2) approximately by the following equation f where C . In Figure 23 the temperature curve in heavy concrete walls is represented as a function of the wall thickness. however.C h2 (Tl . after all moisture has evaporated. as is also a change in thermal conductivity above and below 100°C. In Figure 26 the conditions are represented for two walls One wall is twice as high as the This must which are similar except for their height. be pointed out.deflection h . and above all on the The mean bulging of unfixed walls can be estimated . 3.material constant f . the bulging generally The magnitude of recedes. After the fire. a difference of 1% by weight in moisture content results in temperature differences of about 25°C on the unexposed surface. height of the wall. The bulging on one side. and in Figure 24 the same curve is shown for a 15 em-thick wall of aerated concrete.temperature on the side away from the fire 2 From this equation the important fact is realized that under otherwise equal conditions the bulging increases with the square of the height of the wall. aggregates and bulk densities.

a Owing ceiling above the burning area is rapidly heated on the underside. gaps may occur in the joints at the margins owing to bulging of the walls. i. increases the length of time of the resistance. but which do exist nonethele~s. steel reinforced concrete slabs Whereas walls are predominantly compressively stressed. or in other words over-reinforcing. the usual height of a storey. The time during which a steel reinforced slab resists the fire is determined substantially by the following factors: a) existing steel stress b) steel covering c) nature and thickness of lining underneath d) quality of concrete e) thickness of the floor f) moisture content of the concrete g) kind of aggregate h) positioning of the slab (restraint) Figures 14 and 29 show clearly that reducing the steel stress. the effect of the greater bulging in fire should be taken into account. In a fire.3 3.e. Therefore. By increasing the covering of the steel. steel- reinforced concrete floors as a rule are sUbject to bending stress.-19mens 2 m to 3 m high. which indeed are rare. in floors of tall buildings there is tension on the underside of the slab and compression on the upper side.31 Floor structures Massive. 3. Besides lateral bending. Generally speaking. because stability may no longer be guaranteed owing to large eccentricity. good 'connection between the wall and the adjacent structural members must be assured in the design. Figure 27 shows the bulges in two walls during fire tests. resistance of steel reinforced concrete floors to fire. In the case of taller walls. the heating rate of the steel reinforcement is reduced and thus the resistance of the steel reTherefore the heating of the steel reinforcement plays a decisive role in the . to the relatively poor heat conduction of concrete the steel reinforcement situated close to the underside in the tensile zone is heated more rapidly to the critical temperatures of steel than is the concrete on the upper side of the floor structure heated to its critical temperature.

more compact concretes are produced. since in prefabricated structures traditional plasters may be used less than before. from Figure 28. however. Priority should be given to this problem in the future. to be sure. which becomes flatter towards the inside.uncoated steel reinforced concrete floors (covering depth I cm) after ref. have been used to increase fire protection. for other reasons as well. For a number of years special coatings. in ref. (25) was made. the increase in the time of fire resistance. condition. even comparatively thin ones. This is especially Today.g. debit side it may be said that in practice plasters are frequently applied in thinner coats than contemplated in the design and that green plasters with high moisture content are quickly stripped off in a fire. in general Since the test described It is therefore easy to understand that in more recent fire tests the favourable effects of plasters are not always evident. and thus shows greater thermal insulating properties. . increase in the time before fracture is obtained only with still greater Plasters in earlier tests showed greater protective An essential The explanation may As a consequence the Moreover. be sought in the fact that plaster has a lower bulk density than concrete temperature gradient in the vicinity of the concrete ceiling surface on the conduction is hindered by the transition between different layers. With these coatings. For a slight increase in the The reason is evident steel covering. between the underside of the floor exposed to the fire and the depth of I cm is greater than between I cm and 2 cm depth from the surface. steel covprings.-20inforced concrete slab to fire is increased. of asbestos. heat On the effect than corresponding increases in steel covering. e. curves may be designated as follows: Curve I . Fi~ure 29 gives two examples showing The the effect of coatings on the heating rate of the steel reinforcement rods and hence on the fire resistance time of reinforced concrete floors. is that the plaster adheres well. for example from I to 2 cm. concrete technology has advanced. the heat insulation can be so greatly improved that the fire resistance time is increased to many times that of the uncoated floors. true if the underside of the concrete is smooth. It can be recognized that in the side of a wall or ceiling That is to say. since the plasters have been detached from the concrete at an early stage in the fire. (25). vermiculite and perlite bases. is not very great. side of the fire is steeper than that in Figure 28. the temperature difference A substantial nearest the fire there is a very steep temperature gradient.

it should be borne in mind that under otherwise equal conditions. a higher modulus of to a B 300 the modulus of elasticity increases by approximately 50%. Perhaps the explanation is that with the increase in Given equal In going from a B 160 This concrete grade the modulus of elasticity also increases. these findings for comparatively low grades of concrete must not be applied to higher grades.the same construction.-21Curve 2 . but with a lime-cement plaster 1. also. however. reinforced with comparable steel rods. cross-sections. in more recent tests. for a concrete floor 12 cm thick with an average reinforcement component. The considerable effect of the thickness of the floor on the fireresistance time is represented in Figure 31. minutes. Curve 3 . thickness and reinforcement the floor. and secondly very compact concretes do not behave at all favourably in fires. Another favourable result is that the changes of form of a thick floor are comparatively less than those of the thinner one. Moreover. These results. As will be shown later. the moisture content of a thicker member is generally greater and hence it takes longer to heat through. the fire resistance times in other structural members. which were not always confirmed are somewhat surprising. and. displaces the neutral axis of the cross-section upwards. poor and medium quality concrete up to approximately B 300 the ability of a steel reinforced concrete floor to resist fire increased somewhat with increasing concrete grades. because here the modulus of elasticity does not increase greatly. and under the most unfavourable conditions tend to spall. since the bulk density and the thermal conductivity of the concrete in general increases with increasing grade.the same construction.5 cm thick.5 cm thick. elasticity of the concrete means a lower steel stress. In earlier tests it was established that for The effect of concrete quality on the fire resistance time has not yet been fully clarified. causes a drop in steel stressing of about 6 to 10% (Figure 30). it may also be assumed here that despite mathematically equal calculated steel stress. are increased by increasing the The influence of floor thickness on fire-resistance time is . This relates to concrete slabs Aside from the fact that comparatively large cross-sections behave more favourably in fire owing to their greater heat capacity. the actual steel stress in a thicker floor is less. With an unplastered ceiling this corresponds to an extension of the fire resistance time by about 10 However. but with a cement-lime-vermiculite coatin~ 1.

before failure of the strongly heated horizontal reinforcement. From a fireproofing point of view. The behaviour of single-bay slabs fixed at all sides has not yet been fully clarified(3 4 . Collapse does not occur until the yield point is exceeded in lower and upper reinforcements. The static system also influences the bearing capacity of a floor in Continuous slabs are more favourable than single bay slabs. the resistance of massive steelreinforced concrete slabs can also be affected by the choice of aggregates. at first the still relatively cool upper column reinforcements act with increased stresses as cantilevering reinforcement. since obviously other factors. a continuous upper reinforcement throughout the bay is especially favourable. In construction. have a very strong influence on the test results. these values are much too unfavouThis gain may be particularly important especially in prefabricated A fire-resistance 3 cm. increasing the steel covering from 1 to 2 a fire. 67 ) .-22of special importance in practice. moisture content at the time of the test. time of 90 min ("fire resistant") is readily attainable with floor thicknesses of about 15 cm. rable. concrete floors of similar nature but made with different aggregates had 65 minutes 2 hours . Besides these possibilities. etc. therefore. As a consequence there is a relieving of the stresses in the bay reinforcements. the available data do not provide a uniform picture. the basis for the pertinent regulations. for example. To be sure. the following fire resistance times: Flint Dolerith Basalt Crushed brick 31 minutes 41 minutes According to Davey and AShton(3 4). Table III gives a summary of the increases of fire-resistance time as a result of the above-mentioned measures. the restraint of the slabs in the experiment. for the thicker floors now being used predominantly in practice. the case of continuous slabs.. The values obtained formed However. because in Germany almost all relatively extensive series of fire test have been carried out on steel reinforced concrete floors of thicknesses less than 10 cm. where for a great variety of reasons plastering on the underside of steel reinforced concrete floors is not possible.

-23-3.32 3.321
Prefabricated floors Fire-resistance time In recent years prefabricated floors have been very widely used. Among these a few basic types may be singled out from the very large number of systems found on the market. 1. 2. Beam floors with intermediate structural members. Ribbed floors with intermediate bearing members or compressed slabs of poured-in-place concrete. Floors consisting of beams or planks laid side by side. Prefabric2ted slab floors. Since the steel reinforcements are frequent-

3. 4.

The behaviour of pr-e f abr-Lc a t e d floors in f l i-e Lr: principle is similar to that of solid slabs. ly better protected than irl solid slabs, in many cases a longer fire-resistance time may be expected than solid slabs of equal thickness. Accordingly the measures indicated under point no. floors.

3.31 for solid

slabs for improving the fire-resistance time hold also for prefabricated Investigations have shown that almost all non-prestressed reinforced concrete floors which were given a coating of lime-cement 1 1/2 cm thick on the underside survive the required test time of 1 1/2 hours in the fire test according to German regulations. In Table IV, examples of typical prefabricated floors are given with the corresponding test results. floor slabs are also listed. For comparison, the results of a few solid

3.322

Deformations Besides bearing capacity, changes of shape in the course of a fire

are also important in the behaviour of prefabricated floors. their shape. downward.

Directly

after the start of fire exposure, floors exposed to the fire begin to change As a rule they bend toward the fire, i.e. they begin to sag Figure

32 shows the deflections of some of the prefabricated
The value of the deflection depends not only

floors listed in Table IV. the material.

on the span and thickness of the floor, but also on the construction and For examole, hollow beams of heavy concrete or solid lightweight concrete slabs bend somewhat more than solid heavy concrete slabs under similar conditions. In the case of prefabricated floors of beams or planks laid side by

-24side, we warlted to know how they would behave in the presence of fire under partial
loadin~.

If the deformation of loaded parts of a floor compared A number of floors were

with that of the unloaded parts, there is a danger of shearing at the senaration joints and the fire breaking through. therefore
investi~ated

in fires, only half of which were loaded on one side All the floors (floors without

with the calculated permissible loads(43).

transverse reinforcement and with relatively large deflection were deliberately chosen for the test) resisted the fire under these conditions and the fire did not break through. In addition, small floor units of aerated As is evident from concrete were tested in unloaded and loaded conditions. the beginning of the fire test. sides.

Figure 33, the results showed that the deflections were almost the same at This means that the initial deflection depended entirely on the temperature difference between upper and lower Only after exposure to the fire for over 20 minutes did the deflecAt this time a gradual reduction of the modulus of elasticity tion of the loaded floor begin to increase more strongly than that of the unloaded. began. loading. In certain exceptional cases, floors are deflected at the start of a fire not towards the fire, but away from it. This happens, for example, if the ribs of a ribbed floor are heavily insulated in order to protect the steel from heating, and the upper compressed slab heats up more rapidly than the ribs. On the other hand, a permanent upward deflection after the termiThis occurs, for example, with As a result of fire exposure the concrete The result is a "prenation of a fire is comparatively frequent. some aerated concrete floors. Even at this time, however, the deflection component due to temper-

ature differences remained relatively large compared with that due to the

after cooling undergoes a slight, permanent increase of volume, while the steel reinforcements revert to their original length. stressing" effect which causes an upward curvature. 3.323 Downward propagation of fire Generally speaking, fire tests are carried out under the obvious assumption that fire spreads upward from below, i.e. floors above the fire are mainly endangered. Sometimes, however, we are interested in the For this purpose, behaviour of floors where the fire is propagated downward. hollow planks laid side by side(44). in the usual way

tests were made with a partial prefabricated floor of steel reinforced The same floor had already been tested The tests showed that the floor (cf. Table IV, No. 4)~

-25remained intact under fire exposure from above during the test time of 1 1/2 hours. The permanent deflection of the floor after the fire was Figure 34 shows the upper side of a greater than in the regular tests.

concrete plank of the test floor after the fire. 3.324 Prefabricated slab floors In the prefabricated, large-area steel reinforced concrete floors, which have very recently come into use in apartment houses and in industrial construction, it is often not possible to apply a plaster to the underside of the floor during the prefabrication process. Subsequent plastering may Under these circumnullify the economy of this method of construction. measures: 1. 2. By decreasing the reinforcement stress By increasing the reinforcement covering By the use of suitable aggregates By incorporation of an ancillary layer of lightweight concrete with low thermal conductivity By incorporation of slabs of heat-insulating materials either in the concreting stage, or subsequently, over the finished and installed ceiling. Simply increasing the reinforcement covering is of limited value. An unplastered high-grade concrete floor sometimes fails because one or more pieces of the concrete surface spall off and at these points the steel reinforcements rapidly become heated. portions cannot falloff. Structural measures should therefore
te~peratures

stances the fire-resistance time can be increased by the following

3.
II.

5.

be taken, e.g. by incorporation of a light wire mesh, to ensure that spalled At the very high occurring in a fire, heat transfer takes place primarily by radiation, and therefore particles of concrete that have loosened but do not fall still provide adequate protection for the steel reinforcements. Ancillary layers are made mainly of lightweight concrete or from slabs of other heat-insulating materials. The required fire protection has also been attained in prefabricated concrete slabs simply with the aid of gypsum boards or cement-bound wood-wool lightweight boards which are merely concreted over (cf. Table IV, no. 2).

e.'floor structure". Opinions differ as to whether the behaviour of steps in a fire must meet the same requirements as floors. slabs and beams is possible. in the case of beams the greater structural height is a positive factor. 2.4 Prefabricated steel reinforced concrete steps Steps constitute a special case of the . From these tests it may be assumed that the behaviour of steel However. since in the case of beams the ratio of the surface attached by the fire to the cross-section is less favourable than it is for slabs. it was again found that the fireFigure 36 shows the effect of In all the tests. i. Figure 35 shows examples of two designs which can be regarded as "fire-resistant" according to the existing German regulations.34. and in others beams of rectangular cross-section.47). The effect of the steel covering has been investigated in considerable detail. have been investigated. Basically the same considerations apply from a fire standpoint for steps as for floors.46). which owing to the different sizes of beams and different bearing arrangements are not directly comparable. resistance time did indeed increase with increasing cover. whether they must remain effective as space separators and at the same time retain their bearing capacity. or whether only the latter is important. i. but not proportionally to the thickness of the covering. it was clearly evident that the increase . 3. they have a fire resistance-time of more than 1 1/2 hours(45. an amendment may be exrequire both specifications to be met. Prefabricated steps are generally produced in the following structural forms: 1. In some instances T-beams. no direct comparison between steel reinforced reinforced concrete beams is determined by substantially the same quantities as thgt of slabs.-263. 4. 3.e. In the case of beams. Large plates with steps mounted on them Beams placed side by side Side walls with steps placed between them Beams with steps mounted on them At the present time German regulations However.5 Steel reinforced concrete beams Test results on the behaviour of steel reinforced concrete beams in fire are available from various countries(25. pected. On the other hand. steel reinforcement covering on the fire-resistance time.

Figure 38 shows the ed in Table V. In Figure 37 the effect of age is reprePresumably the increase of fire-resistance time with increasing age can be attributed to a simultaneous increase of the quality of the concrete and the modulus of elasticity.-27of fire-resistance time is less than the increase of steel reinforcement cover. rapid failure occurs. preference towards spalling is noted. which A . in the slender stems of I-beams and on the under sides of the cross slabs of T-beams. especially in the manufacture of prefabricated steel reinforced beams. very compact concretes. For example. In these members strong heating produces very high compressive stresses owing to the inhibited expansion. With structural members of high quality concretes there is a danger of the explosive spalling If parts of the beam reinforceIt is assumed that this ments are thus exposed. Nowadays higher grade concretes are generally employed. for example. The age of the member influence the fire-resistance time of T-beams more definitely than plates(25). Probably other influebces also contribute to the tendency of concrete to spall in fire. shortly after the start of exposure to the fire. the test results are contain- spalling is caused by water vapour over pressure due to evaporation of the This view is supported by the fact that such spalling occurs predominantly in green and very high quality. of pieces of concrete from the surface. of the beams. In considering the earlier test results on beams it should be mentioned that the beams were made of relatively low-strength concretes. that on rapid heating moisture stored in the aggregate cannot escape rapidly It is probable also that the tendency to spall is influenced by the design of the cross-section and by the mechanical stress during the fire.e. sented. spalling was observed in very This may be due to the fact Probably this has more to do with intrinsic porosity than with the kind of material involved. and these stresses cannot be absorbed even by concretes of very high strength. moist mortars made of lightweight aggregates. As in the case of slabs. i. out. enough through the more compact cement paste. an influence of the aggregates cannot be ruled For example. water present in the concrete. an increase of fire-resistance time can also be obtained by a suitable choice of aggregates(3 4 ) . cross~ection Three beams were produced which differ only with respect to the kind of aggregates employed. More precise investigations should be carried out and published as quickly as possible into the causes of spalling.

With the advance of prefabricated construction a desire has arisen to use uncoated columns with high fire-resistance time. this covering is more severely stressed at the beginning of the fire in columns than in other structural members.Sl-SS). or coating falls off prematurely. so that in the presence of fire there is a greater danger of buckling. since subsequent coating tends to nUllify the economic advantages of prefabricated construction. the resistance of a steel reinforced concrete column is influenced primarily by effectiveness of the covering over the steel.e. quality grades were still comparatively low. large and the reinforcement ratios were low. which prevented the plaster from dropping off in the course of the fire. long fire-resistance times. Therefore.6 3. and in some cases heavily reinforced 34. in the case of reinforced concrete columns the steel is stressed in compression. i. extensive series of fire tests were carried out on steel reinforced concrete columns C48-~O) ). the column cross-sections were Very high fire-resistance times were attained with plastered columns where the plaster was furnished with a light wire-mesh inlay. new series of fire tests have been conducted recently at various places with slender uncoated. Owing to the severe compressive stresses in If the concrete the boundary zone. The compressive strength of steel. More recently there has been a strong tendency in construction towards reducing the dimensions of individual structural members. This has meant the employment of higher concrete grades and larger proportions of reinforcement. is also reduced under the influence of high temperatures. architects have been calling for ever slenderer columns in fafade. the compressive yield point.61 Columns steel reinforced concrete columns Whereas in steel reinforced concrete beams and floors the reinforcements are generally subjected to tensile stress. In particular. In earlier years.-28certainly does not occur on all occasions. The fundamental tests of At that time concrete Ingberg and his co-workers are still indicative. columns C In addition to depending on the cross-section size. 3. then even thick columns will not attain The most important measures for raising the capacity for resistance of steel concrete columns to fire lies in holding the .

limestone or blast furnace slag fragments. the use of certain aggregate materials. increasing the reinforcement percentage. since the steel reinforcements of a heavily reinforced column heat up just about as quickly to the critical temperature as they do in a lightly 'reinforced column. given to equal reinforcement ratio. done. obviously. However. constant heating of the column. be assumed that the test results would be greatly influenced. Figure 9 shows the results of recent fire tests with prefabricated columns carried out under the auspices of the Federal Association of the Concrete Block Industry(53).e. The time until failure is also influenced by the size of the applied load.60 m long with the following .g. Fe where: K b F b cube strength of the concrete cross section of the concrete o = tensile yield limit of the steel st F = steel cross-section e In principle these conditions also apply in the course of heating. No data are available on the influence of the degree of It can. e. increasing Fe' under otherwise similar conditions has not so great an opposite effect. the wire mesh inlay can sometimes be dispensed with. cases. in place while it is being attacked by the fire.-29covering. From the equation the following conclusions may be drawn: With increasing concrete cross-section. is obtained. whether of concrete or plaster. This is best done by inlaying a light wire mesh When this is With between the reinforcement and the surface of the concrete. the altered properties of the material. restraint and the slenderness on the fire-resistance time. columns 3. taking into account. the concrete covering is less liable to falloff. The bearing capacity of a compressively stressed column is described approximately by the following addition law: p In such collapse = Kb . however. i. F b + 0st . because complete heating of a larger cross-section up to the critical temperatures must take longer. the fire-resistance time has increased. cross-sections were tested: F F 15 cm x 20 cm = 300 cm 2 15 cm x 24 cm 360 cm 2 In this. fire test. and hence a more even course of the The following observation may be made in here.

as in the German results. 3. and some of these are plotted in Figure 39 as well(3 In 2). Under the calculated permissible load columns of larger crosssectional dimensions fail later than slender columns. so that the applied load on all columns was equal. The choicE of aggregates has a very great influence. however. Figure 39 shows a definite increase of the fire-resistance time with increasing concrete cross-section. the tendency is the same as in the test with constant reinforce- The most important preventive means is to see that the covering of the steel does not drop off. Another successful means of increasing the fire-resistance time of columns was found to be the displacement of a substantial part of the reinforcement towards the interior of the column. fire resistance times of more than 1 1/2 hour were obtained(55). Figure 42 the results of the fire tests of the columns represented in Figure 40 are compared with the results of more recent French fire tests(5 Since in the latter the test arrangement (column length 2. are more or less proportional to the concrete cross-section. aggre~ates. To sum up. Nevertheless it can again be recognized that the fire-resistance times in the case of a French test. ment component. columns. With heavily loaded spiral columns. Figure 41 shows a column after the fire test and the quenching water test. is the excellent behaviour of thin columns. the results are not directly comparable.30 m) differed. The inlaid wire mesh is clearly shown. The . A wire mesh inlay will Limestone 2. accomplish this. especially when limestone is used as the British fire tests showed a similar result on somewhat shorter 4). In the French tests the reinforcement ratio ~ was substan- tially decreased with increasing concrete cross-section. concrete columns behave better than concretes with quartzitic Basalt and blast furnace s~ags fall somewhere between the above-mentioned groups. aggregate.-30F F 20 cm x 20 cm = 400 cm 2 24 cm x 30 cm 720 cm 2 Figure 40 shows the column cross-sections. which have a somewhat greater resistance than rectangular columns. Striking. Despite the radically altered reinforcement component. and for others a quartzitic material. For some of the columns limestone was used as the aggregate. the behaviour of steel reinforced concrete columns in fire may be represented as follows: 1.

where similar conditions prevail. 4. the example of thin. resistance time. This entails a smaller heat capacity and a more rapid On the other hand. shows that suitable means are available in order to give even . of in a fire than columns of moderate dimensions. 4. and has the effect of reducing the firereinforced concrete columns dealt with above. course. that the concrete grade of the tested column was very low. case of low-grade concretes. In the owing to lower temperatures does not then share the longitudinal elongation of the surface. Given suitable design and a good use of proper aggregates. Generally speaking. The interior of the column. as well as the covering of the steel. The decisive factors as Any means by far as the fire-resistance time of a structural member is concerned are its cross-section and the heating of the steel reinforcements. The reason for this comparatively poor behaviour of the thick columns may be that the share of the heated surface in the total crosssection of the column is comparatively small. the dimensions of prestressed concrete constructions are smaller than ordinary steel-reinforced member of equal bearing capacity. times been made in real fires as well. the bearing capacity in the boundary zone is then quickly exceeded. 3.1 The Behaviour of Prestressed Concrete Structures in a Fire General Non-prestressed steel-reinforced concrete and prestressed concrete have much in common from the standpoint of fire. and heavy loads must be borne by the thin surface. construction to fire.-31influence of the reinforcement ratio is slight compared with that of the concrete cross-section.62 Unreinforced columns From an earlier time we have the results of fire tests on unreinforced columns(SO). which this heating is delayed have a positive fact on the resistance of the The heating of the steel reinforcements is affected essentially by the size of the cross-section and the shape of the member. heavily steelcomplete heating of the members. high fireresistance times can be obtained even for very slender columns. Figure 43 shows that very thick columns may fail faster This observation has someIt should be borne in mind. 4.

is gradually relaxed and finally disappears altogether. The steel stresses are not proportional to the external moment. 2. Tn order to attain a maximum fire-resistance time. neutral axis changes its position. This results in severe overloading of . are lower than the calculated stresses. with steel in relation to the decrease in high temperature yield point is less favourable than for ordinary steel-reinforced floors.-32slender' CJ'():. it is necessary. as well. owing to changes in the modulus of elasticity of steel and concrete. the compression zone of the concrete. In many in- stances the actual steel stresses in non-prestressed concrete This results in an extra 4. When this happens The the reinforcement then acts simply as a non-prestressed reinforcement. Figure 14). This happens in non-prestressed floors. The critical temperature of the reinforcement rods in the case of prestressed concrete floors is somewhat lower than for steel-reinforced concrete floors. and owing to the creep of the heated steel. Characteristic of the behaviour of prestressed concrete floors in a fire is a very marked more rapid increase in its deflection before fracture occurs. The drop in high temperature yield point of the prestressed reinforcement rods is greater than that of ordinary structural steels for non-prestressed reinforcements. 3.s-:~cctI on s a relati vely h i gh resistance to fire. but is not so readily recognizable.. and ultimately results in failure. Therefore a reduction of the external load is not so favourable as in non-prestressed concrete. safety factor with respect to fire. of course. The bearing parts of prestressed concrete floors are generally produced in the prestressing bed with immediate bond. The shapes employed are similar to those of non-prestressed prefabricated floors. The reason is that the ratio of the working stresses of the Moreover. The reinforcement cods are of higher quality (cf.2 Prestressed concrete floors Prestressed concrete floors are usually prefabricated parts. The different types may be used either with or without additional poured-inplace concrete. progressive heating the prestressing. Different dimensioning procedures are employed. The main diffcrerlces between prestressed and ordinary steel-reinforced concrete f'r-o:n the [Joint of view of fire technology are as follows: 1.

A slight increase in the fire-resistance time can also be obtained by "over-dimensioning". In Table VI we represented two typical prestressed concrete floors which satisfied the required fire-resistance time of 1 1/2 hours in Germany ("fire-resistant" construction)(37). where the failure of unloaded floors this is because in the case of the unloaded floor the compressive stress due to prestressing and thermal expansion on the side exposed to the fire was too great. the fire-resistance time of a floor under small load is somewhat increased. . lightweight concrete below the An effective way of doing this has floor beams. in prestressed concrete designs the external moment is not proportional to the reinforcement stresses.56) . There have Presumably been exceptional instances.31) and also by suspended membrane ceilings. and hence premature splitting-off of the reinforcement covering occurred. In A considerable improvement in the fire-resistance times of prestressed concrete floors can be attained by means of special plasters (cf. This increases to be the application of an ancillary layer of supportin~ both the heat insulation and the adherence of the plaster.-33for the above reasons. heatin~ of the prestressed reinforcement rods in the fire as long as possible. to delay the been found. Suitable coats of plaster or suspended membrane ceilings can result in very long fire-resistance times of 4 hours and more. 2. means ~an Either of these increase the fire-resistance time up to 4 hours and more incertain designs. As already stated above. because during relaxation of the prestress the instant at which tension appears in the compressive tensile zone occurs somewhat later. for exampl~. occurred earlier than in similar floors subjected to a load. 3. (52. The decisive factors governing the fire-resistance time are the dimensions of the cross-section and especially the thermal insulation of the reinforcement covering. the following can be stated about the behaviour of prestressed concrete prefabricated floors: 1. The critical steel temperature is somewhat below that for nonprestressed reinforced floors. Many other floor tests are described . Nevertheless. of course. Section 3. To sum up. which is of great importance in the attainment of a high fire-resistance time value.

it is heated through more rapidly than a slab which is exposed to the fire on one side only. Since in the case of a beam the heat attacks from several sides. 4. and the TT-cross-sections which are common in America. they nevertheless furnish very valuable information. 7. among others. the U. 6.67) and in other countries(60-63. Although agreement in all points is not found in the various test reports. 4. as in the case of steel-reinforced to heat it through. for these forms also it is true that the fire-resistance time increases with increasing Table VII gives a survey of the cross-section forms of steel-reinforced concrete beams investigated. Most fireproofin~ measures and modifications have effects similar to those exerted on non-prestressed reinforced floor construction. Basically.59. however. Form and dimensions of beam cross-section Covering of reinforcement rods with concrete Nature of the bond Kind of aggregate Load End restraint Additional heat insulation of surface Causes of spalling 8. 3. The higher the desired fire resistance.S. but primarily also on the covering.69) on the behaviour of prestressed concrete beams in fires. 2. Rectangular and thick I-crass-sections behave very similarly. the greater must be the amount of concrete covering . the less time it takes For this reason. With flat beam cross-sections. cross-section area. on the behaviour of prestressed concrete beams were investigated.3 Prestressed concrete beams In recent years there have been numerous systematic experiments in Great Britain(57). The rate of heating of the steel reinforcements depends not only on the beam cross-section. it is less easy to eliminate the size of the cross-section as a parameter for the fire-resistance time.-344. The smaller the beam cross-section. In these tests the following influences. the Netherlands(4 7). 1.(58. the size of the cross-section has been found to be an essential factor governing the fire-resistance time. 5. More difficult to recognize is the influence of the shape of the cross-section. concrete columns.A.

steel coverings listed in Table VIII are obtained ~ccording to ref. taking into account the beam crosssection. The author is unaware of many systematic investigations of the influence of other aggregates such as limestone or blast-furnace slag. It is very important. we do not have adequate information. With small coverings.24). (47) for the various fire-resistance times. Reducing the external loading has a positive effect. Section 2.1 this influence is not nearly so great . for the reasons explained in Section 4. slender beams the covering must sometimes be still further increased. In British tests(57). either. In comparing beams with instant and delayed bond. about the influence of end restraint on the fire-resistance time of prestressed concrete beams although in practice such restraint is almost always present. With very high. The values listed in Table VIII apply to compact cross-sections. which was very effective in the case of columns. on the whole. as in unreinforced concrete(47). The values given are based on the assumption that the covering From a comparison of Figure 44 and Table VIII va1u~s will not falloff and that the concrete is of a technological quality such that no spalling will occur. it can be inferred that the safety factor. resulted in about 20% longer fire-resistance times(59). The investigation of a lightweight concrete of bloating clay with about 2/3 the bulk density of the heavy concrete investigated for comparison. of Table VIII already provide a certain reason longitudinal bar and lateral tie reinforcements must be applied. especially in the case of larger beams. This is because the high cement injection mortar in the prestressing channels has poor heat conductivity (no aggregates) and a substantially higher specific water content than the concrete. This water content delays the heating (cf. In Figure 44 we have the results of several tests The influence of the The minimum on steel-reinforced concrete beams with instant bond. Escape routes for the water vapour should therefore be provided.-35of steel reinforcements. were obtained(47). To be sure. beam cross-section and the reinforcement covering is clear. As in the case of non-prestressed reinforced structural members. similar results. the high content can also have a deleterious effect if the vapour pressure becomes too great and causes concrete to spall. an increase of fire-resistHowever. where the longitudinal expansion was prevented. beams prestressed after pouring often showed somewhat better behaviour. For this that the concrete covering cannot falloff during the fire.

According to American investigations(58) the effect of fixing depends decisively on the magnitude of the "restraininR. This influence is AccordinR.1 Chimneys of lightweight concrete pipe sections Unlike the structural members hitherto described. for example. The Behaviour of Other Concrete Parts and Construction in a Fire 5. the temperatures of the In domestic heating plants combustion gases. In addition to these "normal" temperature stresses. 5. are considerably lower temperatures between 200 and 400°C can be expected. represented systematically in Figure 45. considerable quantities of shining soot in suitable conditions can be ignited. but also in the course of normal operation. Certain fuels tend to deposit chimney. but a decrease for large ones. This soot under t~e in chimney fires and so-called "burning-out". in the case of space separating structural members of prestressed concrete the increase in temperature on the side away from the fire. for example. of vermiculite concrete.-36ance time was found for small beam dimensions. When chimneys are being burned-out. still higher temperatures can occur in chimneys. e. According to Ashton and Bate. This is the case. . becomes the determining factor for the resistance time. encasing a beam in vermiculite concrete 25 mm thick increases the fire-resistance American tests show(57) that the fireresistance time of a TT-beam is doubled by spraying on a 12 mm coating of vermiculite and tripled by spraying a 25 mm thick coating. not the collapse of the member. however. to this curve the shortest fire-resistance times are obtained for total end restraint and Between these two extremes the bearing capacity under fire stress can be prolonged in such a way that. This can happen either involuntarily. force". or can be brought about deliberately by the chimney sweep or the fire Different views are entertained on the effective duration of these high temperatures(6 5).g. Obviously prestressed concrete beams can be very well protected against attack by fire by means of an additionally applied heat insulation layer. time by more than 2 1/2 hours. They probably jepend very strongly on the specific conditions. especially of the fuels temperatures of 10000C and more occur(64). department and kept under control. chimneys are under stress from high temperatures not only in the catastrophic case. than those that occur in a severe fire. which are brought through a flue. To be sure. simple supports.

and then. Figure 46 The above-mentioned requirements are determined by tests which differ in execution from the normal fire tests described in Section 3. because they are subject to mechanical stress both during fires and when being swept. They must be sufficiently stable. In any case. In recent years chimney pipe sections of lightweight concrete have proved very effective. In normal operation the outside should not get too warm. multiple-shell units have been found more effective. In order to meet the contradictory requirements of good heat insulation and adequate gas tightness. Section 3. must not exceed a certain width. For small cross-sections generally one-piece units are employed which may be of either solid wall or cellular design. Chimneys must have adequate mechanical strength. conditions during burnin~-out are very similar to those arising in a fire. sheet 6 "heating plants-testing principles for domestic chimneys". a narrowly restricted range of bulk densites. and Mohler(65). 2. In a chimney fire. porous lava slags are employed. As an aggregate for lightweight concrete chimney pipe sections. after cooling to a burn-out test. chimneys have to be burned-out fairly often it has been found useful . For larger crossWhere sections. bloating clay and broken.-37employed. 5. depending on the material.1). must be observed. They must be sufficiently gas tight to prevent any combustion gases from leaking into adjoining rooms. shows the temperature-time curves for heating and burning-out tests according to DIN 18160. so that undesirable phenomena such as the discolouration of tapestries. preferably blast furnace pumice stone slag. First a specimen is subjected to a heating test. A number of requirements must therefore be laid down for chimneys from the point of view of fire protection: 1. etc. will not occur. so that no ignition by transfer will occur in adjoining rooms. 3..1. 4. Systematic tests have been carried out by Seekamp Many chimneys made of such pipe sections have also been Hence any cracks that may occur tested on behalf of individual companies. the temperatures of the outside of the chimney must not exceed the limits laid down for space separating structures (cf. crushed brick.

The deformation of the wall. satisfied the requirements laid down in the German regulation.2 Glass block walls Glass block walls occupy a special place in the field of wall construction. Figure 47 shows examples of several types of chimney pipe units. and from this to determine the fire-resistance time. or concrete roofing tiles. whereby glass bricks must withstand standard fire for one hour and then withstand the quench water tests. is clearly evident in this picture. asbestos-cement slabs. In a fire these must remain space sealing~ No requirements This wall has are stated for the rise of temperature on the side away from the fire. which after cooling is restored to its original shape. Mathematical Determination of the Fire Resistance Time Attempts were made at an early date to determine the fire resistance of construction mathematically in advance. what has been said in Section 3. For example. tiles are among the so-called "hard roof coverings".-38to incorporate a thin reinforcement of annealed wire in the chimney units. The values thus obtained and the heating through curves are very interesting.3 again applies. This . sufficiently resistant to air-borne fire and radiant heat. to be sure. Figure 49 shows a timber frame building which is veneered on the gable side with asbestos cement slabs. 6.3 Roofing Concrete roofing materials consist predominantly of: lightweight concrete slabs. i. Figure 48 shows a glass block wall in the course of a fire. The latter two products constitute roof covering Asbestos cement slabs and concrete roofing Both types are materials in the narrower sense. Busch(9) attempted to calculate the time taken for structural members to heat through in a fire with the aid of the familiar general Fourier differential equation for heat transfer. and give many indications and suggestions. they protect a building from the effects of neighbouring structures on fire. but they describe the actual conditions in many cases unsatisfactorily. 5.e. The building is completely undamaged. For the case of lightweight concrete slabs. 5. although a neighbouring building has been burned to the ground.

For the behaviour of a structure in a real fire. the rate of heating of steel It can be assumed that with detailed This would apply. On the basis of these magnitudes. The Behaviour of Concrete Products and Prefabricated Concrete Parts in Real Fires In the foregoing sections we have dealt mainly with the behaviour of the various concrete constructions in the fire test under standardized condjtions. e. for example. for 1 hour and 50 minutes or for 2 hours and 10 minutes. for example. prefabricated concrete parts. what matters. In the Netherlands. it is of minor importance whether a given structural member withstands the standard fire. are only conventions. The order of magnitude is In any event it makes little sense to the author if a large This is precisely analysis of the test material at present on hand. to the reinforcements has been suggested(4 7 ) as a basic parameter for the behaviour of reinforced concrete structures. so that for example the moisture content may have a decisive effect on the behaviour during a fire. a sufficiently accurate advance estimate of the fire-resistance time should be attainable.-39is mainly because the material constants change radically with ttle temperature and generally they are difficult to take into account.g. it should always be borne in mind that the required fire-resistance times. Only through experiments under strictly defined conditions However. gates. like the variation of the fire with respect to time as assumed in the standards (temperature-time curves). a number of other essential size and shape of cross-sections. of the kind of aggre- number of constructions very similar to each other have to be tested in extensive experiments in order to satisfy a standard. etc. what has happened in recent years in the field of concrete products and 7. At the same time. influence factors can be eliminated. but to draw general conclusions from the results of fire tests on various structural members. the static system. Also the conditions under which a test is run can vary greatly. . Recently attempts have been made to deal with the behaviour of constructions analytically not on the basis of Jrathematical theorems. it is the behaviour of constructwas it possible to investigate systematically the influences on the ppsistnnce of structural members.

to have adequate thrust reinforcement and adequate anchoring of the reinforcements. Generally it may be stated that almost all measures which increase the bearing capacity of a construction under normal temperature conditions are even more effective against stresses due to fire. This can be realized particularly after fires in older buildings from the early days of concrete construction. in which the present-day building code had not yet been applied. To ensure this. should not be too small. The great differences of stress occurring in the concrete in a fire are especially critical at points where the concrete cross-section is reduced. in fire disasters. quired.-40ions in a real fire which in the last analysis really counts. the elimination of influence factors is more difficult with respect to real fires than for fire tests. By investigations on burned buildings it has been established(66). in beams and columns sufficiently compact longitudinal bar and lateral tire reinforcements are reFigure 50 shows that if the stirrup spacing is too wide the reinAlso. not be so clearly observed. A comparison must Considered this rather be made primarily in terms of the general behaviour. The falling-off of the covering is also favoured by a narrowly spaced supporting reinforcement. especially in older buildings. The observance of the minimum distances between reinforcement rods laid down in steel-reinforced concrete standards is therefore of special importance from a fireproofing point of view. will be in exact agreement in fire tests and real fires. a number of characteristics become evident which in tests could . the diameter of the ties. the propagatioll of a temperature through a construction. To what extent do the results of fire tests accord with experience from real fires? From the very outset it cannot be expected that detailed findings. It has been found important. whereas if the tie spacing is more compact it columns. since otherwise the considerable deformation due to the high temperature On the other hand. practical experience also shows that the covering of the reinforcements is of particular importance in all steelreinforced concrete structural members. To be sure. Like the fire tests. stays on.e. way there is far-reaching agreement between tests and actual fire results. i. that the thickness of the reinforcement covering is less important than that it be prevented from spalling during the course of a fire. at and between the steel reinforcements. especially in the case of forcement covering falls off. for example.

In fire tests. were tested as statistically determinate bearing structures. 8. adequate disposition of expansion joints. To sum up it may be stated that for concrete. in this respect. In large buildings it has been observed after fires that such large mutual displacements of entire sections of a structure took place as a result of great changes of shape. however. Throughout this period it has been found over and over again during fires that concrete is a building material that is outstandingly resistant to fire and is.stresses cannot be absorbed. it has been found that slight lateral support or restraint produces a considerable increase in the fireresistance time. During a fire For this reason. prefabricated concrete constructions. It has the advantage over combustile materials of being incombustible. on the other hand. Furthermore. More favourable in many cases. the bearing capacity of concrete constructions in fires in a large number of cases has been found in practice to be greater than in the fire test. there is a certain lateral support as well. but which in practice are statically indeterminate. are certain concealed static reserves which do not show up in the test. its large mass and its high specific heat. that the individual structural elements were no longer able to absorb these displacements(68). where in general free lateral movement is permitted. and it compares favourably with metals by virtue of its relatively low thermal conductivity. has an unfavourable result. This applies particularly to constructions which This is the case for numerous In practice. steel-reinforced concrete and prestressed concrete constructions there is good agreement between behaviour in a fire and behaviour in a fire test. Failure to take into account generally benign construction principles which are of great importance in the case of a fire. concrete and reinforced concrete structures are not able to resist fire indefinitely either. One effect which is not covered at all in fire tests is the size of the building. Summary Concrete and reinforced concrete construction has existed for over a century. rivalled by hardly any other material. The latter are almost always tested as monoaxially restrained. in all buildings exposed to the danger of fire there must be an . Despite these favourable conditions. especially prefabricated floors.

Furthermore. material properties of the concrete and reinforcing steel. By taking advantage of the knowledge gained in fire tests and in fir~ fire damage to buildings it is possible also to obtain considerable bers favoured by that method of construction. these dimensions are. stability conditions are affected. Depending on the dimensions of a member. generally speaking. materials and methods. insulation afforded by the concrete cover to the steel can be improved by plastering or by special selection of the aggregates employed. by means of suitable technological and constructional precautions.-42- the structural members affected are gradually heated through and through. the distribution of forces within the cross-section and sometimes also the Since the penetration of the heat occurs more slowly in members having larger cross-sectional dimensions. this heating process proceeds at a higher or lower rate. resistance periods in precast concrete construction with the slender memThe great spread of concrete and reinforced concrete construction over the past hundred years has occurred not least because . reinforced concrete and prestressed concrete structures for any desired fire-resistance period within the limits encountered in actual practice. It can safely be assumed that in the future. in the case of reinforThe thermal ced structures. this superiority will be retained with respect to new building .these materials are superior to other building materials in the event of fire. the fire resistance is very largely dependent upon the protection of the reinforcement against excessive heating. the most important factor affecting the behaviour of concrete members in fire. With progressive heating. to build concrete. To summarise. changes occur in the In addition. it can be said that it is nowadays possible. too.in addition to other advantages .

/Okt. S.7. (2) l\. [40) Perstlnllche Mlttelulng del' Firma Rhelnbau GmbH. 15. ten Eisenbetonbautellell und Stelnelsendeck:en". Hami lto'n. V. Sonderheft. (13) Lehmann und G. 1930.istisch cs Jahrbuch rur die Bunllcsrepublik Deutsch· lnnd 1962". Washington.. "Fcuel'beton auf POl"tlandzement-Basls". Bericht Nr.".untl Werkstoll. Wlesbaden.M. Heft 60. 1962. [l9) C. Journal of the Amertcan Concrete Institute. Wilhelm Ernst & Sohri. "PrUfung der Feuerbestiindlgkelt von Baumaterlal". H. IngoHg. "Uber die HeiBdrllckfestigkeit von Beton". Kristen. Stuttgart 1957. "Fire on Structural Elements" National Building Studies. 4 (1963) Nr. Hermann und Wedler. (5) K. Vol 43 (1943) (34] N. Dusch. Ludera. Paris. "Verhaltcn dcs Bctons in nonercn Tempcraturen". (23) Personllche Mitteilung des Fachverbandes tur Hochofenschlacke. "A Short History uf the St. [10) S. Ashton. (29] H.:Ober <las Verhallen des Betons und Elsenbetons Irn Feuer und die Ausbtldung von Dehnungsfugen Im Eisenbetonbau".Erhohung der Feuerwlderstandsdauer von Stah1betonrippendecken durch Verwendung von Holzwolle-Lelchtbetonplatten". Archiv tur Et!icnhUttenwescn.ics or Concrctc at High 'pompcr-a tu rcs". g ta t. Wlesbaden 19(i2.. A.. (20) E. Blunk. 17} H. "A. ]9~. (1I] H. 1938. Berlin. Forster. 161 K.f.e comportement du be-lon entre 80 et 300 0 C. Berlin...St. "Das Verhaltcn von Dcckenkonstruktioncn aus Stahlbetonfcrtig-tcllcn bet Brandvenuchen". Heft 122. [15J L.. [:i7] Th. Berlin 19. Neuwlcd. Malhotra. Sept. Grijn und Bcckmann. Berlin 1911. Mag of Corrcr. E.t. (24) Htitte. 1961. [8) R. Civil Engineering and Public Works Review. 102. Seekamp und W. 1!1l8. Vol 8.S. Kurzber. Berlin. Heft 89. Dcutsche r AU5. "Heat-Resistant Concrete". Fire Research 1957. Reihe D. Tonindustrle-Zeitung und Ker amtsehe Rundschau Nr. T. (12) K. Res. R. Baulndustrle Helt 2211959.B.A. ".Feuercinwirkung aut nicht brennbare Bausfoffe und DaukonstrukUonen". Mall.vcrstencrunuswtrtscha rt". Gesundheitsinuenieur. "Die Wlirmeleitflihigkelt von Beton In Abhanglgkeit vom Raurngew ich t und Feuchllgkellsgrad". D. Heft 8. Mehmel. [26] Dannenberg. L. ncrun. Man 1962. R. 23. Vol. Zurich 1941. R. 12000 C". PInkston. Deutschmann und Melchior. . Gn r y . 17. S. 1907. T. S.Fesllgkeitselgenst·haften del' St"h1e bel hohen Temperaturen". [3] . A. Wallo. Nck rassow.. Menzel.ig. ~3 und 41. (22) Personltche Mitteilung der Firma Deutsche Siporex GmbH.s Stationary Office.tlnllche Milteilunt: des Verbandes RheinischeI' Blmsbaustorrwerke e. [25) Kristen. 1959. DUsseldorf. 6. 5. Auflage. . (18) Woolson. P.-439. 5. [21] O.. Bau u.tcr Einwu-k ung von Temperaturen hi:. (14) A. Investtgauons on Building Fires Part. Holz ats Roh. H. Cruz "An Optical Method Cor Determining the Elastic Constants of Concrete". huB rur Slahlbelon. "Some Physical Propc rt. London 1953 [35J G. 5.B. zemenrKalk-Glps. D. Essen. (4) . [391 Perstlnllche Mitteilung del' Firma Lenz & Co . [33J C. Kriseher.. Munchen..'Ire Resistance of Walls Lightweight Aggregate Concrete Masonry Units" National Bureau of Standards BMS 117. London J9SS. Kohlhammer.und Gcf(ig-c~inderung von Bc tonzusch laustof Icn und z crncntrnortctn ur.. 84 (1960). (9) H. . Taschenbuch rur Eisenhtittenieute. g'tu ttunr-t. 1931.'·. l'hillco. 1930. April 1961.ructural Fire Protection of Building!'. Hasenjll~er. Herr Mayest)·. Bautechnlk 1/1962. 12 Her Mayesty's Stationary Offlce.. (36) Per. Bauvcrlag GmbH... Meddelande 50. Fortschritte und Forschungen Irn Bauwesen. (16) L. c Wilhelm }':rnsl & Sohn. Heft 27. W. (27] Ros und Eichinger.HltzebeslancHger Belon".. Diss. Braunschwe ig.l!ang Nr. "Die wlssenschafllichen Grundlagen der Trocknungstechntk". Aurlage.. D. Strey. Davey und L.. Kristen. 138. zemcntverrag Gn:bH. 1935. . Raisch. . "Problemallk und Systematll( des Verfahrens zur Prtilung des vernaucns von Holz und Holzwerkstolfen im Feuer". References (II S.A. V. 3. Jahr. No. Springer Verlag. D. (1'JJ -t.f. Deutscher AusschuU fur Eiscnbcton Heft u. Research Paper No. (38J Pe rsorrlteh e Mitteilung dcr Firma Verelnlgtc Baustolfwerke Bodcnwerder GmbH.. auue. Bornemann. EMPA. D. 1916. 14. Enrtcll . Bautorscn. Stockholm. Seekamp. . Journal of the PCA Research and Development Laboratories May 1962. Becker. Proceedings". Herr Mayesly's Stationary. Ludern.v c rsucbc tiber Lannen. (31) H. "Brandversuche mit belaste. (30) H. 78-79. Berlin 1929. . April 1958. London 1958. Batlr No. [32) Schlyter. . . Th.Baulicher Feuerschutz. "V.Drandprobcn an gtscnbc tonbauten".tarmzerreIBversuche mit Sparmatahleri". Werkstoffhandbuch Stahl und Elsen. "The Elleet of Temperature on the Compressive strength of Concrete". [28) Ros und Eichinger. 1956. .trntcrsuchungen uber <len EinfluB von Temperaturcn uber lOOn C auf dl~ Druekfestlgkeit von Beton". "Tests of the fire resistance and thermal properties of sottd concrete" Proceedings ASTM.St.T. Omnla-Decken-Organlsation.

Stu turart. Japan Society or Civil Engineers. H. or Civil Enginet-rs. {OOl H. Heft 5.-J. Bat". IBD Burcau of Standards \Vashington 1921. Sch ulz c-Wcdter-. PCA BuI . 13. cccdmus of the Symposium on Prestressed Concrete and Composite Beams.Publication SP-5. (45) r-crsonucne Mitteilung der Firma Betonwerk Karl BUrkle. Thoms.M. Ingbel'J.ahtbetonbaus". Seekamp.Fire Resistance of Concrete Columns".nco: of Prestressed Concrete Civil Engineering an Publlc Works Review Vol 52. Herslellung.. Deutscher Ausscnun fUr Stahlbeton.Das Verhalten von Deckcnkons tru k t i on cn im Feuer bel Brandausb reitu ng von oben nach unten. Berlin 1956. Cnh ict-s du C. JUst.-J. C.BI'andversuche mit stark bcwehrt en Stahlbetonsaulcn"... Fackler.. Berlin 1959. 8. Wilhelm Ernst & Sohn. . 15B). A.is de Rc~istancc au feu de Poutrcs et pfanchc rs «. H. . Deutschcr AusschuB rur Stahlbeton. [46] r'crsonucne & Co .Brandversuche an Hausschornstei. r'ersonttche Mitteilung des Bundesverbandes der BeLonslelniud uatr-le.957. (HI Th. "The Fire-Resistance 01 t-rcs trcsscu Concrete Beams". of Prestressed Concrete Slabs". " }o~5s. Deutscher Ausschun fur Stahlbeton... Robinson Wilson. Versicherungswrrtscnatt" lB. (62] Th.U. 1591 C. (47) "Brandproevcn op vooruespannen Betonliggers". Junl 1960 [UI TIL Krislcll. Wilhelm Ernst & Sohn. Stahlbctonbauteilen und Stah lstef ndccke n bei Brdnden". (66) ]\1.-44- (-{IJ Pt~r:-. Smart..S.« Essaisde resistance au i'cud'elelnents de Construction (Be series)». Juni 1960 et No 49. Kristc n. P. GustaCclTo and C. 44. (1963).• Mittcilun~ der Firma Heumar. (G1] Th. Detroit 1962. Beton. Carlson. Skokie 1961. The Fire Resist.lstllll~~r~illl~ bci Brand vcrxuche n an Deckcn aus nouenctnandcrttcucndcn Fertigbalken oder Platten".. Ashton und C.. Heft 92. C. nen".T. (631 K. . H. 1/)000. Lon. "Tcmperatunnessungen belm Ausbrerinen von Schornsteincn'\ warrncwtr tschart. K. Mohler. C. Bnk kc. VC1'wcndunl: 13. Oktober l!. tcttns 142. J. "Fire Tests or Building Columns". C. "The Ncw tjcam Frn-nace at PCA and Sornc Experience Gained frorn Its Usc". "Sponsored Fire-Resistance "rests on St. [6B] C. . Kawagoe. Wierj~•. Kristen. (65) H. Betonstein-Zeltung 5/)959. Novem. "An Interpretation of Rcsut ts of Fire Tests or rrrcs tresscu Concrete nUilding' Components". H. Ir. 156) A. Jahrgang (1935). (55) H. [42] J. Gl'lscnkirehcn. PCI Journal. Seck amp. No. "T'he Fire-Resistance of Prestressed Concrete: Concrete & Constructional Engineering. H. Norues nranntekntske taborntortum. bel' 1955. A. nlpl. Berlin 1959. L.i)llli('h(' Mitteilung dcr Firma perrckra-naustorre Miescn Co. HcuCers. Shulcn .. R:lpport Nr. Heft 7/8. don 1960.& Ih. . 1. P. Cahiers du Centre Scientlfique ct Technique du Batiment No. Heft 132. Bc rf ln 1939. Ashton.. Tatm"lllll.. J.tjn torsuehunucu uoer unatj nstlue . Wilhelm ErnSl & sonn.. Hannemann und H.R. 2 .l6'l. A. Der B. lngberg.T. Technical Papcr Nr. Carlson "Fire Rcslstance of Conerete". "Fire TcsLo. C. Knoll Koln.ructur-at J:::lements".Widerstands(ahigkeit von . Rapor t {4B] (49j (50) (51) (52) (53] [~41 C. "Brandversuehe nut belasteten Eisenbetonbautcttcn". No 617.illingenieur. Heft 132. A. . Amsterdam 19~8. stark bewehrten Stahlbctonsaulen hoher Betongtite". A. wird dcm- nuchst verorrcnutcnt. (57J L. W. Hill.-J. "Der Jo:inCluU holler Temperaturen au( Bau tct le aus Spannbclon". Wierh~.lrboll an P. (f4) Th. Technical Paper 272. November 1957. wasntngton 1925. April 1961.Drandversuche an sehlankcn. Wicrig..'I. Her May~sty's Stationary Office isso.. (67J C.l: Griffin.. J. Tell II.. Kristen. L.-Ing. July 1959. Bureau of Standards. Ashton and P. Reiter "Das lo'euerrisiko des industriellen St. Jahrgang Nr. Hill. C. Kluz. ACI . l-'nrkl c r... "Brann{orsk mcd vanug arrner-tc of for-spent armcrte betonjetkcr". Pro.B. .

40 Philleo water 28 days * Testing at 110°C.26 Age at time of test 28/50 days 28 days Endell Nekrassow Prism 7 days in water then air 18 x 18 x 100 mm wet* . L = 100 mm prism 51 x 70 x 152 mm cylinder 0. prior to the drying.-45- Table I Test conditions for the measurement of thermal expansion of cement paste Author Storage before test Dimensions of specimen Water-cement ratio 0.0' 16 mm. .

5 plaster. boards olastered R"".. ~f'f- . ~.. cm Material Bulk density R (air ary) Compress.-46Table II la.':: '~ ~ ~ ·Slftm . 1...5clD R • nO kg/1Il 3 a Dr • 35 kg/c1ll2 F ::: 1/2 1 1 1 '4 54 40 e) e) I I 1/2 1/2 57 26% by wt.7% by wt.. test finished 1 _ r. slabs 120 x 50 x 7. test finished .a Dr Moist.. l% 22 j oint 3mrrr:::~.cont. 650 ke/1II 3 1/2 1 1 1 38 44 a Dr not tested F = 1/2 1/2 60 12. Wall crosssection dimensions. F Temp.ern lime cement mortar (S jOi~~ 5 (" 1 V t% ~ ~ ~ I Pumice cement boards 100 x 33 x 5 cm R..7% by wt. inc... I Ref.cemented L~ ~ aerated conc... : 2 3G cjr~" '"' ~ '" F? ~ 1 r:0 Pumice cem. on side away from fire Test time (h)" 6T in Mean o( --- 1 I 2 I 3 -5t- I ~2 ~ 4 I 5 I 6 I Max I7 I 54 54 74 I I 36 ~ P.. test finI I ished I inadequate *) I formation of joint .650 ke/cm 2 1/2 1 1 1 21 29 24 34 54 a Dr not tested F = 1/2 1/2 45 12..str..

1/2 7 35 R F 1 1/2 15 39 16 47 54 . 5 35 no fines conc. blocks 24 x 12 x 10 cm 6 35J" 1/2 1 = Dr 1310 kg/m 3 1/2 6 9 36 = 66 kg. 3 3 70 -42 test finished slag conc.-47Table II Continued aerated conc./cm 2 2 21 not deter. crushed clay brick 7/15 mm R 1/2 8 1 1 2 = 1330 kc. 2 3 50 test finished ./m 3 20 45 a Dr = 46 kg/cm 2 1/2 1/2 36 test fin- F not tested ished solid slag conc.5 cm R = 550 Kg/m 3 a Dr = 2) kg/cm 2 F = JL2% by wt. blocks 4 22 50 x 25 x 7.

nd c...ravel R = 2200 kg/m 3 0 ..itic sand 3./ / f- heavy concrete quart:.0.... 5 - F' not testerj test fin-ished ..""'44 ""71 ""80 - v0 V~/~L z-.63 Dr = 310 kg/cm 2 3 ... 33 ~ v..Continued II - a 34 9 33 10 33 - ~o- . 2 1/2 .-48- Table . 2DOk!}/m 3 rJ/Z.

perlite or sprayed asbestos base applied underneath Increase of concrete grade from B 120 to B . 40% 10 to 30% 50 to 100% 200 to 1000% 5 to 30% approximately 45% .225 Increase of floor thickness from 10 to 18 cm Additional time before collapse in % approx.5 cm plaster of strongly adhering lime cement mortar Special plasters with vermiculite.-49- Table III Increase of fire-resistance time of massive concrete slabs owing to design modifications Modification Approximately 30% reduction of steel stress Increase of steel covering from 1 to 2 cm 1.

::.". ../ :/(/ ~. H +' 0. ." I Gypsum board slab with fibreglass added Cone.. Ref.... 70m 1/2 h 1h 1 0. I Q) Q) DC o <+-t+.:"Y'in~ :22m c o Ll a o s e Cone. .1\ steel.+' (Ij Behaviour in fire test .. (Ij .. \ .o8~ 8 -r-l >:: +' ·rl 8 +' [fJ'O Q) >:: 8cD >:: 0 +'>... s tr. ///:.. pY 3 maF~B. o:r../ /~.....('ij .1.'>. massive slab I 2 38 :..o rll Q) <+-t'O 8 'rl +.. str....:X/.3.// r .05 70 133 205 7 23 63 simply supported M.\X:~1:" /?A0:~~~ . Floor design Dimensions in em 0..:H8 ::J [fJ [fJ 0 Q) 0.. -----8 Q) (jJrl.610mkg ""15 = >-3 cr I Vl I-' (j) 0 I < H coV:..".~ // ' < <> /. IV..·rl ..35 0.a : 528 kg/CIn l ~ mkg 1/2 h Fire test concluded - I I . ~Yer lll. : WJ6 '260k9/cm l Heavy cone. I \ I ? 4 I lh 1h 5 1/2 1h : I 1~ 6 I 7 305 475 545 I 8 48 90 94 ill 1 I (: J..0 . r t "" 2.a-ld-.(:. ...//~.:o<+-t H 'rl 1 I 2 I Heavy cone.s ~ /-.000>..65 1.c ·rl No. +' 00>:: >:: Q) 'rl 8 H 0 (Ij 8 Q) >:: +' cD Mean temp..• J/..//{.."./.L 15 J <:>.Q)Q) 0 <+-t >::+' 'rl >:: H o Q) Q o:r.."'" >::(ljoo >::8'018 cD '---" >:: '0 '---" Q)'rl -----.'" ~T I\L f5. ~ '\ I . U ·rl cD Q). . 65m simply supported M .~ 34 ~9~~ -"I steel ~~: ~ I t l e . . /'..

98 m 1/2 l" Conc. plaster Fire tes 1 concluder .5 1.!270 mkg (Feldmoment) 4 Fire test concluded ~6.neavv concrete . \ .::: (]) I~ H f..62.7 Fire test concluded I.4 Fire test concluded poured-in place r--- !.~.5~m e.2 21' 5 39 simply suppo:-oted If.2 1. 3 34 ~·{~.05 9..00 m '.~".00 m 1/2 4..-" '54 48' n 104 I~ >= 1 1/2 6.1=unknown '92 55 cern. t = 3.1 5. str..: W so:28Dkglcm 2 ~f fixed all sides crossed reinforce- '4 5' 69 125 15 29 50 91 2 mlf.. ..00 m <: o o 1/2 I Vl f-' '.. I IT .= 4.6 1/2 5. = 4.5cm lime '.EI 45 J'.'. 4 4.1250 lIlkC '5 ::::.15 28 37 simply s uo o or-t ed --33 6' 82 M_960 mkg '/2 I 0. ::::.(/{~t~ ~~ .'las s·iY_~-slab ( ~: < :~(:: : .0 155 296 29 6 37 simply supported .

mesh 8 22 '/. ::::.red-." pio L <:. (lJ t 9 I 41 = 3../~ Vt: ////// 0 v r' / / ~. lIJ /.85 m 11 1/2 1 I 0.00 III 1/2 1. .3 3. (1" H I Ul fD --so I Steel cQverinq.P"".--/' / simply sUDDorted IvI=unknown .."vs("Inj SIQbs o: (lJ f--' I\ \ \ 1/2 7. s tr.75 3./fJ.9' 26 46 ~-3 !OJ ' ~_ _-.= 7 40 4.:207kS/cm2 I . (.3 47 94 146 365 simply 1 supported 1 1/2 3 M = 2120 mkPl Fire test concluded I I 13 30 Cone.85 • m 1/2 1.I'"r.4 1/2 2..9 1 . ~ " f-". "5.75 Fire test concluded < o o ::::. ~.1 21 44 = t 280 mke ~------~-e ~lT1ent p!Q-5let'" I'"rt.\-~e.-. >= c.5 0.-qTed c. 1 53 98 124 11 s Lmp Ly supported l~ 1. e=3.

(28) No.-53- Table V Influence of aggregates on the fire resistance time of steel-reinforced concrete beams after ref. but about 6 cm crushed clay brick concrete on bottom of beam Leighton Buzzard sand + Torphin Whinstone (limestone) 3h 32m 3 more than 4h (test discontinued) . Kind of aggregate Fire-resistance time hours : minutes 3h : 03m 1 2 Alluvial sand and gravel like 1.

7 em "H 1 1/1 2. Ibelow: 9 steel rods 3x8mm sl 145/16 37 t.n dimensions in em Span Bea rin'6 Bending mOrrEn at mid-bay rl (mkg/m) l (0"1) Behaviour in fire test '='est time in h Deflect-I Mean temp. 4. Ref.5 a '.No.r:::I I- .opposite ment fire I Reinforcement above: 4 twisted wirf's211t5moY\~150HSf.0 em 70 -< H einforcement t = 4. Floor desiq.OOal 1/2 2.5 em 72 PJ (j) 0' f-' I \.OOm 2 1/2 0. in DC ion at id-bay ]at reat side ~nforce.25em 51 ~ simply supported ime cement plaster 515 K • 1640mkg/ml 1 1/2 7.3 em ( 172) 41 lime cement laster light wt .]1 ..9 em ( 93) ( 10:5> 13 37 simply sunported K ~---62.1 15.

r°-.l....2 . r.8 12.2 20 210 " 2"'. No 1 IRe/47 I I I Shape of crossdi"lensions in C'Tl ... 1TI-: . 4 S. " . '. (." .- -- 0 ooj 150 150 50 40 45 " ..0 r- 0 .I I r . - 10 15 I.5 G. ~ e'f- I 119 25. 7 6' l~i 1 L " 4 15 24 '.5 8 6' Ijl :1 I .4 to tc 22'.l. . t: ~- 5. ".7 ----.4 2.6 to 50. 50 lH.4 - .jei- 14O 18 11. -' .( '2." -c .\ _CII.2 )0...Q to to '6 - I - J. b I c I d .-55Table VII ..'- I --"IJ" -. .. .~ // _~-Q---t ~' • /_.5 14 14 6 'if~ J[f: .~ II I .6 to to )0. _ _ _ ---'--J Co T 25.2 120 '5 6 '0 11 -~(.0 g' :<. _.0 14..5 G. :"0 oJ ! I . r . I I I '0 2' I I i i 0 2 .0 ito 60 to to 20.J. /. .4 to .' _~----L to In to to 78.1\ .~.5 1'.ll lfr rill I.5 . .4 12.

0..uncertain due to evolution of gas ~ig. .1956 1955 1551.. I. "'I I H >:: .._ . 1. BosolI. I r:' .) QJ 0 H (Ij I 1:1 QJ .... J I .. Sondstcin. .. 3.r::: . Kolkstein. [/) --1I . Aglamere..Unsicherer Verlou' infolge Gasobgabo Fig... limellonej 4. . Grauwacke.. agregalS lolan (5). Fig.. ....l.r! ~ >=: 0 >=: 3 . 1 0 9 DM "~ ... . 2. Basolle. r1 (Ij .... 2~- 200 --1'1'1 P -".. 1954-1961 Bild 2. . 2.furnace .. . Diagram showing 'jre damage in the German Federal Republic in the years 1954-1961.. par incendies dans 10 Republiqu8 Allemande duront les onn. 6.. groywacke. 2.. 1. (Ij ~ QJ .. Linear thermal upanlion of various pggregate5 according 10 (51: 1.. Lineare Warmeousdehnung verschiedener Zuschlog~Ges'eine noth (5). Brandschadensvorlouf in der Bundesrepublik Deutschland in den Johren 1954-1961 Fig.. 6... 5. 4. . 6. b05al.---"" "-- ~ 5 1959 1551 I.r! ~ >=: -1 0 200 ~oo 600 800 Bild 1. 1. I I m]Gross social ~l product in '100 r. 100 - Ii :1: 1 I 0 I . . broken brick.-56Table VIII Minimum prestressed reinforcement covering values after ref..e. --- . Ziegelbruch . Gr'" 3. . 2. 1. I".. Lailier de haul fourneou en morceau . (41) Desired fireresistance time hours 1/2 1 1 1/2 2 Required minimum re 1forcement covering in cm for beam cross-sections of 100-200 200-500 500-1000 1000-2000 2000 3 2 5 "--------- 2 4 6 4 6 2 3 3 4 2 3 3 5 8 10 500. - 1557 1958 . Tuile. 2. Hochofenl'uckIchlocke. Dilatation lineaire par 10 chaleur de diver..j.. concanee. Degel.Determination incerloine d" degagemen' del gal a CO". crulhed bla.. Colcaire. S. 1960 -- "--I--- I j--- I :liDamage due to iHire in 10 6 DM 300 .Iagi 5... 10ndSlonej 3. 4...

4.i de._ - -_ .. land evcc diyors agregCJIs lelan (5) Prapar'ion du melange 1: 3 : 0.. 1. 3... 2. Dilatation lineaire par la chaleur de mortiers on ciment Part. ><: <1> f1 E--l .. 4.J-i-r-1 i i. lincar thcrmal expansion of molars mode with Portland cement and various aggregates according to (S) MiJC proportions 1 : J : 0. Rhine grayel... 'Ion COl Fig.froidilsementl lelon N..kallow (6) l '0 . I -. according Endoll (5) . 4. Warmeausdehnung 'ion Zemen'stoin b. Grayier du Rhin.. Lincc re Wormeousdchnung yon Mort. land) chouffe pour 10 priCmier. hil:rung {Erwarmung und Abki. 2.1 200 .n g according.~ I i I" . . Linear thormal axpanlion of hardened cemont pOlte (Portland cemont) on baing healed for tho .. linoar thermo I cIpanlion of hardened cement pOlle on boing hooted for the second time (heating and cooling) according to N.iten Er. durci (cimen.. rl (1j -1 H 8 \1-t-"i'~1~--'" ! --~)...froidiu8ment } Echouffemen. 2.... Linoar.rstmaliger Erwarmung ... S. chauff' pour "n.... limestone. Kolkstcin.67 noch Gcwichtsteilon 1. laitier de haut 'aurneau en marceauit'. HochofensIljduchlaclr. 4. 5. men..o.00 H 8 E--l .. Linear..jrst limo.-__ • -.---t-..Philion 171 Fig..Determination incerlaine 0 caule du degagement du C02 \ 200 400 600 800 1000 1100 Tempe"alur ·C . balalt . _ Ph. 80solt ...nnch Endoll (5) } .. Philloo (7) " 0 c J -- ·rl UJ >=: 2 (U P..lar" ( ><: <1> rl (1j H 8 .. Rhcinkiesel.. dvrel.67 en poi'dl.uncertain due to eYal"lian a' COl Fig.-57- "C-..00 600 Tlimperatvr 'C Temperalur ·C Bild 4.! i r--I I J...~-t:-.• -- solon Endoll (5) R.Unsicherer Verlauf infolge Abgab.-. Fig.. 3...ihlung} noch Nekrouow (6) Bild 3.. Dilatation lincaire par 10 chaleur du cimen. Warmoousdehnung 'ion Z.ln aus Porlland.chooli.... sec ende foil (echauffomenl of r.. und vers chiedencn Zuschlogstoffen noch (51 Mischungsycrholtnis 1 : J : 0. aw.i . _ . .. ! I I I I I % o~ 1\.67 by weight 1. 3.Abkiihlun g ell N 0 k fOllOW (6) _ _ __ Erworm"ng nO _ .---1-----I 800 rl (1j o 1------ i I 1000 1..ement) b.luoslOW (61. 3.-- I-'-'h. \. 3..o NQkroliow (6) _ cooling Fig.c <1> .... Calcaire. Basalte . Port~ foil $ 8 I N k rauow (6) on 8 . Dilatation lin60irc por 10 chaleur du cimon. o '" I : c UJ '2 ·rl s::: (1j 0.c p <1> 0 ---- v"-: .lIoo (7) Fig.c <1> -1 200 600 800 1000 f200 1.. lump blal'·'urnace Ilag: 4..mentltein (Porllond:r. s: ./ ~~ \ \ If \ H -1 (1j s::: ·rl <1> .::l -2 " \ Bild S...

1.froidi..chauff..".. EinfluO hCiherer Temperaturen auf den 0----0 6 t:.1 2010000 kg/c m1 Tcmperalur ·C E ... . 1.. Druckf. rwid 340 000 kg/cm J E..elul.. ComprGllive strength of concrete 01 0 function of the heating lemperalu r• Fig..nl it JOO" C 100') C Echowff. ._ W.".slion dWi b. aft....it Eka ll 400000 340000 240000 250000 290000 von Bclon kg/em' kg/em' kg/em' kg/em' kg/em' (19) (19) (181 (181 (191 fig.I.1 0 W/Z ~ 0.000 kg/cm 1 _ ...(....-.I~W.." un • o-...60 +00 500 bOO BelOn avec colcair. .in·S. -....- +-' UJ 80 .111 • 1 290.ing lubi.sligkeit von Boton in Abhiingigkeit von der Erhi'zungstempera'ur Fig_ 6..000 kg/c"..-.'.... froitl 250 OO!) kg/cm! E . .--j 10 *0 S:::+-' s:::'U Q)H U) t 200 '000 600 s::: . W:Z = eow/cimonl Eka ll = E il frui. do lemperaiurci .. I.. 10 campr.." . froid 400 OO~ kg/c m1 ~ 2..n Ausgongsdruckfeltigkeit 246/kg/cm l Prufung in heiBem ZUI'and _ _ _ _ _ _ Prufung nam vorherig. Milchunguerholtnis 1 : 6 n.... - 20 H 0 +-' 0 +-' (/) 0 2 ... Inilial compreuiv.ra'ure d'. for leverol w.c c. r.\ fruid 290000 kg/em J .<: Erhitzung auf 100 0 C =.9.e ~ 0. Fig...1..'on apr•• louminion • . . WiZ 200 300 0.Iovee... strenglh 246 kg/cm 1• teated in hot condition _ _ _ _ _ _ hllted after cooling healing to 1000 C :-·1 '-" heating to 3000 C Fig.. .raturb..6 w.-_ .z o 100 r..men.000 kg/cm1 00 w. RGsiltanc.d temperotur.40 W..ur 10 module E du boton....ek•• Mix proportion I 1 : 6 by weight.'on E .. .. • .-58- % 120 Q) Q) fOIJlC--.r b. .-.t.limel..--j s::: bJ1 s::: 60 Q) cCrl Q) .40 E"o ld 340. .. B..._ . Gew.l UU'Ut'''.. 1.in-S.-.ctod to vat.000 kg/cm' 1 \ .men.r Abkuhlunlll t..'on Ojobos-S. .. 0 100 ---'-..c+-.----t:.· o "'" •• mphlllhit....llien 246 kg/e".. ..e ~ 0.11 alo D ta compr.pru.--j 0 1-- .aft. .d tempero'urol upo" the modulus of elouicity eE) of coftcr. Resistance Q 10 compression du b. I.·T. 6 8 fa (2 eating time in weeks 800 1000 fZOO Bild 1. Itthluenc.. "'" Echauffem....'.. a M"I.. 6.. Effect of elevol.' .andUone concrete E•.. Ei ' n ld 400.40 E.menl U lild I. "-.'. +-' 00 UJ H +-' ct: oO.lone eencrete E"'''d 240.."tlttn' tIIW.000 kg/em' dleb ese concrete E"lIltl 250.'on en fanction de 10 temp...60 Kolk"..Z ~ 0.on Sand.i frui.i1. Erhitzung ouf 300 0 C 0 rc 0 Iemperatur ·C o Bild 6. Compressive Itrength of concr. Beton ovec diaba.1 _____ Euoi a I'elot chaud E"ai apr'l r. __ Tension ..• "f1uence.60 . chung.Z ~ 0.on E~Modul Eka lt Ekn it Ekn ll E....c *-1-' ..... Oruckfc"jgkcit im Beton noch mehrwochigor Temp.

I I / I7fF IT / ~8 r< >. . 0 aero ted Concrete R ~ 506 kg/m l 1221 6.4 >.j.- / ~~ .air dry -. 9.. Transmission de 10 chaleur ). --Bild 11.midi. !hormol conductivity )..m.) 0 1.'on 5~homoll. 6 fire-cloy concrolo R ~ 1670 kg/m' (6.onduc.j.. Fig.. ~or":.. -... I"h". _ .beton R = 506 kg/m' (22) R = 1670 kg'ml (6) R ~ 1153 kg/ml (2J) o Fig. Coefficient de transmission de 10 chaleur du belon en fonetion de 10 tcncur en humid it.. ·rl c o o 116 "" v. 11.. o D. kiinsllich gelrocknol Fig. Warmclcilzahlcn von Beton in Abhangigkeil von Raumgewicht und Fcuchllgkcilsgcholl bci Roulnlemperolur nach (20\ _______ fcucht densi tv 510kg.-59-- kca! m.) :> ·rl I o ::l o 0.8 / / /" / -c c: a o rl ':::l l/ I m H 0.0 I ·rl ..re. .. / / / 1000 . .6 118 1..> ..C I:-< zoo "00 600 600 fOOD Temperotur °c ."" k / ~. of an aerated concr..0 water-filled pore space total pore space Von . a function of temperalure.empcroturo.j.2 0.8 keol ! I I I I I cs --~-b~ .I ~ ~V I o de 10 tomporolurc 0 Bclon-go< Bolon ovec Chamolto Beton avec ponce do loi. 8' Icche 0 I'air lache or.j./l·"C Hcol m·n··.- I / .'0 1. 0 -----..-Belon Go. Warmclcilzohl von Beton in Abhongigkeit o 6. 0.. aines Gasbotons in Abhongigkei.. 9. a' a function of humtdlty and tomporat".2 8 I:-< 1400 1800 2200 2600~mJ .c <lJ o i 2DO 600 o Bulk density 8ild 9. Ilgkelt und Temporalur noth (:ll) Fouche .~..-----~ X X m H <lJ 8 a....elcitfahi9kcit ...nb. Coofficient do transmission de 10 chaleur du baton en fonclian t:.506 ku!m' (22) R = 1670 kg!m' 161 R = ·1153 kg/m' (231 . Hull.ier R .lof ltrockcn . 10.ivity of concrote o. a la temperoture ambian •• do 10 denlil8 ••Ion (20) humide Fig. according to (:ZO) mailt Bild 10.e of d.t.b. VOn der Tcmperatur 6.8 / /i 1/ .ifici.. 10'.6 0.) ~6 ·rl .llement ~ 0. 0... Thermal conductivity values of Concrete 01 a function 0' bulk dcn:oity and moil'uN conlcnt at room . . d'un beton-goz en fonction d. 11. X X foomed Ilog concrote l R - 1153 kg/m (23) Fig...) rl / -: t.r:::7 0. 10 temperature..) :> 1/...!. Thermal I'.artificially dried Fig./ o t2 10 0.: / .

St 60/90 (/> 26 mm.allod Fig. te ne ur en ciment I Q"". .p JO G ·rl b O O"ri (]) QJ ~ . . choud (0.sO I.c ·rl . Warmltreckgrenze (0.c 0. . 1 100 ~t 60190 \ \. . ..... .. 5.p UJ ~ I G-< 150 0 0 " ~t 16Q/180 0.-: ~t---I I I I V BOO fOOD .-:-. 14. Fig. Lou of .. 12. lo""ni. 12.g:'m 1 :.>.0. nl \V.-60- QJ. Tor sle. Wormedohnung einel Stahlen mit 0.) *!llmm 2 200 UJ UJ (]) ... cold-drown .• . B St I a 8 mm . Linear thermal expanlion of a ...~\.180 <t> 5 mm. paidl du betan avec table •• Browier (oleair."t' 'II" .. . .• . Porte d.r-~"": 300 Bild U..40 ~Fjg.und Spann" stahle B 5.1'r.. 13. Gewichhvcrlult yon Beton aUI kolkholtigom Sand und Kie' b...4'/. . ... p..c u. Torstohl (/> 8 mm 51 160.p rl UJ <l UJ fO Total water 1/ -cant..i Erhitt:ung ZcmenlgehClII Z = 440 kg/m 1 Wauen:ementwert W/Z = 0. kaltgozagen -.. wormgewo!z..'Z -I.... . 16 mm d.)I. modo with coicorooul lond grawel on healing.2".4".p ·rl ~ o 0.lire 'roid .. maud a a 200 __ ~ 400 500 SOO Iernperat ur 'C . allemands pour beton el belon precontroint'. 0 I +.5lru(lurol sleel B 51 I.2 Oehngrenze) deuhcher Betan.. . . Hot yield point (0.~. G-< I ~'rl (]) s::: 0 ~' / / 100 JOIJ 'HIf) 5{)() f(J(J 600 700 800 Tempera/ur 'C Iemperotur ooo °c Bild 12._.15 UJ o. Dilalatian lincaire d'un aeier avec 0.. 60/90. r---.4'/.... (\J ~< '\ \\ 100 0 \ .oeier Tor " 8 mm .. 5 mm dio.2 limite d'elllienlibilite) d'acier.. C Fig.. 13. I <t> 8 mm ..51 160/180... 50 'd rl (]) -" -. Limite de duclilile t. Lineor.oof stren) of German reinforcing Iteell ond prestressing steeh..Stiiib ---~-..• St 60/90 " 26 mm.. lor. at begin~I ning of the I~I I-I~+ --1:. .• ha'·. 8 mm dio. C {24} fig.a .8St.p 05 bOl>'!. e. • . 8 mm dia • • • _ •• .• --. -li. C (24) (24j Fig.. de I·ochouff~m('nl.~+- / k-': ! . 14.t. Comen.eight of concr. . t~ ----I. .tee I containing 0.lm .40 BiJd 13.51 160/180 " 5 mm.. . (ontcnl C""" 440 kg/m ' Water/cemont rotio W/C -.

.mp~ra'ur -C Bild 17. 15. . 0. . von der Tem .. coul.e d'ocien allcmondi pour bOlon precontrainls ci 20 v C oprel i'edloullemenl ci rouge (26) • -.Fil en ocier ftoid pour belon procontrainl SI 160/180 de 5 mm .• Warmgewall. o Q 0 100 '-.2 mm dia.i... -I-' '-.! 100 200 300 400 SOD 600 Temperatur ·C .__.2-0chnprobe deutscher SpannslCihle bei 20U C nod.2 euai a I'extensibili. U..ibili'e jUlqu' Q 10 rupture d'uR Geier coul. <.s in Abhiingig". EXlen. E d'u" Dei".j 'd ~ o 100 200 300 400 500 600 T. Cold-drawn prestressing wire. Fig. >-.ler Siobsiohl S. r= .Hot-rolled bar.-61- kg/ m m ' 200 '/. 100 200 300 . 10 temperolur•.. SI 60 190.::1 ~ -P ctl 20 +) fO >-. . lanl lilieG en fane lion d. Modioli.j °1 rl .2'1.90 de 5._--_. Bruchdchnung oines FluB"ahl. Ultimote strain of 0 mild . lS.hen (26) • -. ctl >-.iliceoul mild steel 01 a temperature.. O. temperolur. 0.00 500 600 Annealing temperature DC 8ild IS..-< . vorhorigom <il. r-l o SO -------_. . -I-' 30 UJ -.. 17.IS annealing (26).teel 01 0 function of temperature. 60. ModUlus function of of ola •• icily (E) of non-. . 10 . Bild U.• Geier en baguette lamine 0 chou'" S.. 5.2 mm cp . Fig. 50 -« -r-l c M UJ1S0 UJ QJ .Kaltgczogcner Spanndraht SI 1601180 von 5 mm cp Fig.. SI 160/180.2 mm a ... 60.90 von 5. proof lest for German pres'reuing etccls 01 20 0 C ofler prsviQI. en 'onetion d. Fig.Modul van Tomporalur unsi'iziorlem Flu611ahl in Abhangigkoil von der Fig. . peralur 16. etire a I .r! -I-' UJ ~ o 2 QJ Cr-<+) » 'r! tr: .. Fig_ 17. 5 mm diD. OJ UJ <t--. E.

mp. -.-62- Q) H ..S".~. Anglete"e.~de a . C (221 01 0 C----Q Pu.~ ..rotu.noch.. C (22) en fonetion Duration of test Gild 19..hland (-I Nicdcrlande. pOyl Allcmogne (-) Pays Bo.. C 1221 USA - -.ition). Gormany (.temp. -.. Land. Fig.rature curv.emperal"re-temps ollvmonde aero bienl" odapt'e S. Thermol conductivity of Iteels function of '. 19..' de.. I I I i I I ! ! Q) ! ./) I ! 300 I i 420 7 O. -.Ocutsc..00 r.m'uion d. curves of varioul coun'ri.?j~.rolur 0 .. F. .--__. i.C 4-t 800 0 S n P..0 .~o .. ISO (Voroehlog) D-----o . ISO (Propo.. I:. s:: H 200 ~-.i'onnien..pualur 'C Bild 18.".. WCirmelcitzahl von Siahien in AbhCingigkeit von de." Jc 10 temperalure Fo. is SOon to be odop'.-- Suisse i:1 .oino.:S fOOD 0 . Coofficien. Frankreich. Ei••n (18) I:. Greot Britoin. GroBb.d to the ISO curve Fig...-one...p qj .p H s:: 0 : 60 f i 120 fOO 3 i 21. 4-t kcot m·h··( ·C 1'-OD 0 s:: 1100 U) Ii! I 80 r-------~--___. France. Q) . pur (18) O. 10 chaleur d'Ge'. I:.S".. Fig. Time-tempera1ur.c:. deulsche Einheil••emperalurzeitku""e wird d. 18.•• Z • 5 3611 6 480 B Min h O. 600 Q) ! .- Schweiz Schwodon ISQ-Kurwe angepaBt (eJ Di... ISO (propOlol) USA Swit'lcrlond Swedcn (-) The German standard time.) Holland.on 1181 6-----1:.:). Temp. 18. - Fig. 19. USA - -. Ligncs de temperolUre-IGmpl de diver. Tomperalurzcitkurwcn wcrschicdene.S". de tron.

duration and bulk density R for a maximum temperature ril.rhohung yon l4DO C auf der dem Feuer abgekehrten S. (diagrammatic I Fig.ling of floor.. Fir. in Abhongigkeit von BrClnddauer und Raumgewidat R 'ur .. Fig. Temperafur. Parol des sni pour 10 res istcnee au feu d'une hauteur d'u" e.ide remote from the Fig.ion moximo du e6le oppa" au leu . Epaisseur necessaire d. 20.. mallives en beton en fonction den.t.--. Woll 01 storevhiqh lightweight concrete slabs prepared for the lire te. 8randkammer lur Deckenprulungen (Schemal Fig.. 21.ite Fig.tonwand. gescha8hahen Leichtbetotlplatten.':<: o 1::~5 ..-63- Bild 21.in.r! +J . Zurn Brundveesvch vorbereitete Wond au. Fig. de paroi.. la durie du feu et de la de la lemperalure de 140° C Oft fire. thickne.5 QJ the . test chamber for Ih.oge.I"'""':I ui QJ . Erforderliche Dick.. 21.en plaques de baton leger em 20 rJl r-------------. maximal. 22.. 1. of solid concrete walls as a function of fir. 20. mOlliver B.it' R pour une augmenta. Requilit.c: rll0 rl cd ~ Bild22. of 1400 C '07. ChoMbre pour I"euoi aux p~afandl Bild 20.r! >-i ::s 0' ~ QJ 0::5 1\ 2 4 8h Duration of fire . 22.

onwand (R 900 kg/m J ) im Narmenbrandversuch... Fig. Fig.:::l cd H QJ .(J cd ~ 115/J .. curve in a 15 em thick aerated concre'. Temperature CUnt. (R = 900 kg/mJ) a I'euai ou feu narmalise. Y2. 23.. 10 temperature danl dOl porois a be. in dense concrete walls of various thicknesse. 2 and 3 hour.c .paineur•• a I'ulai ou feu normali.denor Dicko im Narmenbrand nom V2.:::l 100 H . teneur en humidile environ 81 / 1 .jJ QJ 800 1000 .-64 rl rl ·C ·C 1{)(.~. Temperaturverluaf in einer 15 em dick..- C ·rl .2300 kg/mJ) verschi.c .. Propogalion do la 'emperature danl une paroi de 1S em d'epaineur en beton-ga)'.jJ .~a 8 Bild 23. EO: QJ 0 5 cd H QJ 250 0. wall (bulk density 900 kg/mJI in the standard fire test. 1. 2 ef o 7 II ~5~ Distance from the wall side exposed to the fire 3 heur•• Bild 24. Moisture content appro•• 8'1..two 8 Pro . as obtained in standard fire after Y2..L. 1. 24.jJ 600 ·rl ~ +00 QJ SOD QJ H .jJ 0.n Galbe.on lourd de different. 8 side EO: QJ fa. . remporalurverlouf in 5chwerbetanwandcn {R . . 2 und 3 5.e apr'.n' Fig. Feuchtigkeits~ehalt . 1. 24.unden Fig. Temperatur. 23. rl~~~~~.LJ::Z~. Propagation d.

• Ieinen Wandhohe 3 m (31) Fig.S cm dicke Wand aus leichtbetonhohlblock...istance d'une paroi non encadree de 10 hautewr h. ·~--'I---+----j 3 2 1 2 Duration . tipaineur 7.80 m high (33) . Warping of walls in the fire test . 2'.Paroi en b'ton leger de blocs creux. i / .15 em dicke Sehwerbetonw(Jnd 1. Wall en blocs..80 m (33) . epaiueur 15 cm. 25.. 26... W(JII of solid slog concrete blocks after 0 fire 1est of 3 hours' durotion Fig. Reduction of the "ability of 0 nan-re. Diminution de la re..-65- ". Wood (JUS Sch luckenbatonvol lstemen nach von 3 Stunden Dever einem Brandver.80 m hoch (33) -. 25. 5 -1711 i -_--r:----:--! / I i Bild 27....7.. Abnahme der S'andfe..en Wand mit der Bauhohe h fig.://// ////// 811d 25. Deformation de parais lars de I'euai au feu Paroi en be.. Verwolbung von Wonden im Btand\lersueh "'-.tigkeit einer nicht eingespann.. cm .5 cm thick wall of lightweight concrete hollow block.trained wall with ccnstrectian depth h. fig. 27.15 cm thick dense concrete wall. height of wall 3 m (31) Fig. _ . 1..7. 27.pleins en beton de loirier de haul fourneou apr4U un essni ali fe-v de 3 heures Bild 26. hauteur 1. hauleur 3 m (31) .uch Fig.5 Clft...//// /'/f//.on lourd...

Anstcigcn der Templ!rolurcn an den Bewehrungutiihlen von Stahl· bctondcckcn chne und ".e.nags (l) Platands sans crcpiuoge (2) Cropiuoge en eiment ef chou. 28.iuonee pe. l-. 30..n' I'tlua' • .j. 31.'"..~ .iod of reinforced concrete . Verandurung des innaren Hcbclarms :l infolgc .---~ I j ~ '~600 ~ >:: ~. 30.5 em 8' M/n. Propagation de 10 t. d. Tempcralurve. CJ) :5 I (?) ~I OSOO <+-< -rl >::400 (J) ~3OD / 1/ / / / § 20o lild 28.lauf in .red underside.5 em Zement-Kalk-Vermiculiteputz Fig. plaque en beton Duration of fire test Bild 29.esislance au feu . Temperature rilG' in the reinforcing ban of r..1lm ~ (J) Bild 30. Tempera'ur.". .j. (3) 1..-66~'C o 500 ~OO from thOD E .) (J) "I i i 0 I 8 JOO E.) >:: 50 v: -rl [/J (J) 3D Steel covering Q.< 0 5 10 15 20 15cm de 10 quolile d" butan dans 10' m6mos condilians (Sdt'ma) betcnplouen Slab thickness Bild 31. flu 1:-4 3 -hr- ". (hoRgomant d" bras de lovier inferno z du au changement -rl ~ I'r. plofoRd.inforced conere'.. lomporoturel QUlI armatur. 29. dans "~"'It "lIIt. chou.Y I V -: -: ~ ---:. test fig. 11} unplasterod f100n (2) 1.225 OOOkgllm 2 cd .it Unterpulz (1) uRvcrput11e Dcckcn (2) 1.peratur.' un. Variation of the inlcrnal lover orm 1: dUQ to change of the concreto quality under ofhurwi5e equal conditions jdiogrommotic). sons et ovoc crop. _paineu.S em cement-lime-vermiculite ploster Fig.lab.. vermiculite. SORS' gloichcn VcrhCillnisscn (Schema) Fig. f100n with and wilhout plasl.. EinfluB der Oeckendicke auf die Feuerwidenlandldouer van SiahlFig.. 29. epaiucur 1.inar S'ahlbolonplotlo wiihrond dc~ Normcnbrondes Fig.. 110 (J) I G 90 -r1 ..~ I 1-----J) I E .. p'100 (J) E curve in Q rQinforcQd concrete slab during tho 1// v.. Fig.rmall.(ndcrung de' Be· tongulo unle. 31.. _ _ S ===:.. 28. I Ilondard fir.5 em Kolkzcmcntpul1 (J) 1.__ ' S?.5 em cement-lime plasle. EffClct of floor thickness on Ihe fire .:ume sur 10 .. Fig. Influonce de I'epaiueur d'un plafand en plaques de beloll .5 em (3) Crepillage en eimen. belaR arm.. Augmentation dc".----. 1.

Spalte 1 fjg.l. 35.. 32. Escaliers en beron manufacture fig_ 34.. Die lohl bez. column 1 fig.-67- em 0 10 ..1Q/I Test time Bild 33. ou tableou 4. Bctonfcrtigleile-Troppcn Fig. a I'essai au feu normalise selon (37) Test time Jild 32. colonne 1 Section 8·8 A~ dim.. 33._._.._.. L._.. _ . Brcuduusb reitunq von eben ncch unten...J o 10 20 30 70 80 ~~ 50 60 70 0 10 20 JQ ... Downward spreod of fire...uai au "" normalise.haul en ale tend creux en betcn crme upre s lessoi Ql1 bas. Top surface of reinforced concreto hollow flooring unit after the fire (38) Bild 35. EinfluB der Belastung auf die Durchbiegung einer Gasbeton.. Obeeseite r-iner Stohlbetonhohldiele nc ch dem Brand (38} Fig.I C 0 ·rl +..ieht sith auf Tofel 4.... Propagation dl1 feu de. deckenplatte im Normenbrandversuc. in cm I I At-...---l_.nd en plaques de betan..- Sectiol1A-A ~B ~ Section '8-8 At--- ~-_~oo __ I ection A-A lime cern.. plaster~ jiiii~ :' " " " " " Aild 34.. test..:..I.... Deformation de pla'ands en beton manu'ach". chiffr._..l. a I·.... Influence de la charge lur 10 deformation d'un plafo.:> o ()) H G-t ()) so 6C c:::l 16 L. 32. Cote superieur d'un feu (38) .0 50 60 70 80 901/itl.gaz..'0 30 100 . _...llftber refers to Tobie 4. Effect of loading an the deflection of on aerated concrete floor slab in the standard fire test according to (37) Fig. 33.h nnch (37) Fig. 34... 35. The ftt. Deflection of precast floors in the standard fir. se rapport... Precasl concrele stairs Fig.. Ourchbiegung von Ferligleildecken im Normenbrondversuch.

. . 39.g. o 5 'H o !_--p. I .. . Einllufl der lf bcrdc ckvnq der Stahloinlogon auf die Fouor-W. Oroiccko: Britische Vorsucho . 38. t. Influonce de 1'6go de pcvtrcs en bolon orme sur 10 resi'lof'lt~ au fou Bild 36. .p O 6JO - 1 Q) Ul Ul 500 .-_. em' t I I "1."I -20 _ '00 fJJ-- '-' ~ J QJ..-I ' -1 I '~i :> u o o &ild 37.a .g. diyors agrogats (2S) section des powtres ::J rl U E c o z J 4 hr-s . British tcs ts Dutch tesls German tcSls o X • Fig. f-l0 / l ·rl . Effect of the cover of tho roinforcing resillonco period of reinforced concrele beams.J J --- I ... Essc! au feu sur dolo povtros en plaques en belan arma c. BrancJvcnuchc an Stohlbctcn-Plo ttcnbclkcn aus 'Icrschiedoncn: schlogsloffcn (28).. '6.- ..-------. .: j..5 236.. Cros s-scetien of beom Fig.. M _. I . aggregates (2a).~" 10 resislonce au feu 0 Euais britanniques X Essais neerlandai. Resistonce ou feu do cetcnnos en belon arma rectonguloires on lonelien de 10 section et des agrogals Agrcgat cvec quarlxite • Agrcgal avec colcaire ~lllilC: Deuhcho Venucho.offen !\ 0 quarzitischer Zuschlog A • Kolkstein-Zuschlag f..----. Influence do 10 couvorfurc sur I'ormature en Geier do poulre. 31. 36.:! I c:==:.----------. 2 J hrs.. ~ i I I 1. douor Fig.5 "1 ..dor.-'-:. 13 i . Fire tests on reinforced concrete T-beams mode with veri.u Ei5I _}O r- -.douer Yon Stohlbclonbolkcn 0 britiHhc Vcrsuchc nic dcrlc ndis che Vcnucho • • dculschc Vcnwchc o o en Fig.--------. 4resistance time I I ~" J j j • I in mln.-68- ·rl c bD s. i. beion ormc . Effccl of tho ogo of reinforced concrete beams upon the f.90to " • limestone aggregate ~. 31. <:It ~I .39. rno3.. ime lild 39.p . Einflull dolo Allors yon Stohlbetonbalken auf dio Feuerwidorllond..J -_·_-t~-+_ I I I '3 s...._---. ·rl c: I QJrl ~12.. &0 00 20 I~ uo 1.. Fire rosivtcnce poriod of rectangulor roinforced concrete columns 0\ 0 function of cron·section ond oggregOles 'quorfzitic a99 r. 'tOO 200 r- I Dim. . resistance pcricd Fig. 38.-Ul I _~-20 o t...p . Bdd 38. i I I -- - 'H o ~-r:.-_ _ _ _ _ _ . Bolkcnqucrschniu Fig. • Euoi s allemands § 1000." I I ! -.v.. Fcucrwidcntondsdauer yon rechteckigen Stohlbetonsawlon in Abhongigkeit vern Ouc rs chnitt und Zuschlogs.he firo 0 .- _.in cm.5 ~. sland.. a . ICO I Structural a "j~tl~ rrime V2.ar.. QJ DI1'h inC"'" I·' 1 .-.Clf) ..J -._-.:> 0 ~ ~ ('" . 36. steel upon .0 1..

..60 . agregat avec quartJ:ite ~ = 2_3 J/o. =8cm '" I " ..0.30 m G--i 600I . Soulenlange 3. = 2_3°/0.X ftanxosische Versuche. Reinforced concrete column made of limestone concrete with incorporated wire mesh cffer the fire fesf foltowed by sprayit19 wifh wafer (parf view) fig. Stchlbetonstiitxe au..-69- Seri~1 'r 1 I o Loo p 6:mn e=8cm r Sorie 611.t i r e 20 . Querschnilte der Fertigteihtutzen fur die Brandversuche des Bun~ desverbandes der Betan5teinindustrie Fig.. 1~'/6mm I! ~ ". Crass.---...itischer Zu.t .30 -X-X-X-X Essais frant. Essais allemands. Appui en beton orma..-----::r:-----. Seelion des appuis prefobriques destines cux essais au feu de I'onociotion des fobriconts de betan manufacture.x . . = 2_3°/0. Soulenlange 3...u: quartzitic aggregate jJ. e.i.6--6.------_ . 42.. 4 r r-------.chlogsloU ~..r .....7 S.. = 2_3°/0 j length of column 3.x . agregot avec quartzite ~ = 0. Kolkstein ~l.. Effcct of concreto section and reinforcement proportion upon the firc rosistance period of rcinforccd concrete columns German tCs..".- <------".\ .ais..EinfluB von Betonquerschnitt und Bewehrungsverholtni..Essais allemands. 3 rl >::. limeslone jJ...L = 2-3010... fOr the fire tests of th<l Bunde5verband der Betonsteininduslrie Precast Concrete federation) fig.30 m __ x __ x__ French lests: quartzitic aggregote ~l . / 1"'-.8°/.:> OJ tr: 200 t .German tests..!-ooP . r . 42.8°/0. 40.sections through precast columns. quan. 1-- U BOO 1-----\---- o o Bild 42.. Longueur de 10 colonne 3..30 m Fig.X ..60 m -.5 cm Bild 40. J". 41..teinbclon mit Ma..chinendrahleinlage nach dem Brandversuch und ans-t:hlieBender loschwasserprobe (Ausschn. Kolk. length of column 2. 41..j '0 >::: o .j - o o o ~ 2 3 + hr s resistance time . IOGOr----.. length of column 3. quarzitischer Zuschlog· stoff ~~ = 0. Longueur de 10 colonne de colonnes m m 2.6--6..) (German Steel covering of loops.60 m ."' !.6--6... Longueur d ela colonne 3...... i Bild 41.30 m Fig. Soulenlongo 2.... Influence de 10 section du bdton et de I'armature en beton arme sur 10 resistance au feu.2. I ' i I . auf die Feuerwiderstondsdouer von Stahlbetonsoulen deutsche Versuche. ~ <he s h I.co +. fig.-j.deutsche Versuche. avec colcoire et filet en ocier comma armature ep re s I'esso i ou feu ef ensuite crrcse avec de I'eeu.. 40..F .8°/. Calcoire ~l """ 2-3°/0.30 m ...

> QJ a 0 1 -1..l---.. Bild 45.) 0 ~ I / / " /' ~""..) S QJ m . .. 46. . 45. 44.-----.j..tullo" nam (44) Fig. au feu 6 de' oppuil en beton non arme . Brandvenucho on unbowehrten Boton. immediatement..'armolure de poulrel en b'Ion pr'conlrainl... according 10 DIN E 18 160 Fig.. 43 Ello.... 45.. 44.. ....ail d"chauffemonl el ClU 'eu aur: cheminoel lelon DIN 18160. Wirkung eincr Endciospannung auf dio Fcucrwidcrstondsdaucr von Spcnnbetcn [s chcmcfis ch] [51) F.g.5cm 5 em .. Time-tomperature curve.... ..../O ~ ~ -8 '0 (44) 00 J6 10 ~ O'-------:-:':. . liililil 090 m ~60 -ri UJ QJ.-l ..:. Inlluence de 10 loction ef do la COUYorture lur . . Courbel de tomperature-Iempl pour del e . (God rung one und Quoruhnillo) Fig.j. 43. enu of test ciue ~\ to excessive heat transfer \ ~ ~ . 46.. steel covering ~~ -ri flO _ ~. end restraint p' 'bOY. Effect of end restraint upon tho fire rC5islol'\cC period of prus rros sed concrete (diagrammatic) (51} Fig. . lection. on chimnoy..-.gom Verbund...lectionl) Fig. Bild 43.. 'or burn-au I lelll and healing lelt.. .. 600 '" ~ ~oo oheating C' 200 QJ ~ -ri 0 c.venuche on SchornItoinen nach DIN E 18 160 Fig. : bur'rr-ou t 800 \ test \ UJ -ri UJ 01 ~ ) 0 '.... .. Fire tel" on plain (unrein'orced) (oncre'e columns according beam section Fig.---- ..--. oliomblOo.-70- Hin f5 0 ~ r-. Effoct of beam lection and COYor to Itoel upon tho 'iro rOliltonco period proltrolled beam. lur 10 r'liIionce au 'eu l'e'ite.. with pre-tenlioned tondon •• (Squat lectionl and 1. 5 6 hours Bild 46....elan (441 Bild 44... Temperotuneilkurven 'ur AUlbronn. et en 'orme I) 0' fOf)() ill -. Ein'luA von Bolkonquorlchnitt und Stahluberdodtung auf die Pecerwiderllondldauor von SpannbofonboUren mit lo'orl. .und Heil. .. Effo t d'vnc fixolion d'un betcn precontroint ~ur 10 rthistoncc au feu (Schomotique) (51)...---J ·ri 500 1000 fSOOeml rr.

H."/" "j' :>.-71- L I I ! I ../ /'..hmdard fire le. 47. Wall of qloss building block. during the . . Bcis picle vc n Schcr nsteinfcrmsnickon cu~ Loichtbctcn Fig. Examples of lightweight concrete chimney units Fig. / Bild 47. Wand GUS Gtosbousteinen wahrcnd des Normenbrandversuches Fig. 4ft Paroi en b locs en verre durant I'euai au feu normalise. Bild 48. . Exemple s de pi e ce s en beton leger pour cheminee s.1 Fig. 0 ~ /. 48.

Fig.-72- Bild 49. distances de I'etrier apr•• un incendie (. 49. Verkleidung cus Asbeslzemcnlplalten schulzle gegen Flugfcuer Fig. 49. Reinforced concrete beams with varying stirrup spacing otter a lire (Irom 58) Fig. Stllhlbetonbalken mit unlerschiedlichem Bugelab'lllnd naeh einem Schadensleuer (nach 58). 50.elon 58). 50. Porcmenf en plaques demicnte-ciment protegealr centre un feu emporte. . A facing of asbestos cement slabs afforded protection against flying sparks Fig. Bild 50. Poutres en b"lon arm" a <Iillerente.

Sign up to vote on this title
UsefulNot useful