Thermal Stresses and Movements in Bridges PDF
Thermal Stresses and Movements in Bridges PDF
Thermal Stresses and Movements in Bridges PDF
1972
Recommended Citation
Reynolds, Joseph Charles, "Thermal stresses and movements in bridges" (1972). Masters Theses. 6714.
https://scholarsmine.mst.edu/masters_theses/6714
This thesis is brought to you by Scholars' Mine, a service of the Curtis Laws Wilson Library at Missouri University of
Science and Technology. This work is protected by U. S. Copyright Law. Unauthorized use including reproduction for
redistribution requires the permission of the copyright holder. For more information, please contact
scholarsmine@mst.edu.
CCFSS LIBRARY Joseph Charles Reynolds THERMAL
21 1 * 2765 STRESSES AND MOVEMENTS IN
1972 BRIDGES
Technical Library
Center for Cold-Formed Steel Structures
University of Missouri-Rolla
Rolla, MO 65401
GAYLORO
THERMAL STRESSES AND MOVEMENTS IN BRIDGES
BY
A THESIS
UNIVERSITY OF MISSOURI-ROLLA
1972
Approved by
This thesis has been prepared in the style utilized for publica-
of the ASCE.
iii
ABSTRACT
that the task is extremely complex. Some correlation has been made be-
ACKNOWLEDGEMENT
TABLE OF CONTENTS
Page
ACKNOWLEDGEMENT ~ • i v
INTRODUCTION. .. . .. .. .. .. .. • • . .. • . .. .. . .. .. .. . . .. .. .. .. .. . . .. .. .. .. .. .. .. .. .. • . • . .. .. .. .. .. .. .. .. .. 1
AIR TE~ERATURE.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. • .. .. .. .. 6
BRIDGE TEMPERATURE......................................................................................... 8
CURRENT SPECIFICATIONS......... • • • • • • • • • • • • • • • • • • • • • • • • • • • • . • • • 24
EXPANSION JOINTS..... • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 27
JOINT SEALANTS...... • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 29
FUTURE NEEDS................ • • • • • • • • • • • • • • • • • . • • • • • • • . • • • • • • • • • 35
S~RY .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 36
APPENDIX I. -BIBLIOGRAPHY. • • • • . • • • • • • • • • • • • • • • • • • • • • • . • • • • • • • • • • 38
VITA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
1
INTRODUCTION
short term daily temperature changes and the more lengthy seasonal tem-
are the thermally induced movements, along with creep and shrinkage
due to the lack of a rational design criteria, the design engineer can
not be certain that a structure is both the safest and most economical.
under-design.
I
Graduate Student, Dept. of Civ. Engr., Univ. of Missouri-Rolla,
Rolla, MO.
2Numerals in parentheses refer to corresponding items in Appendix 1.-
Bibliography.
2
effects were considered more frequently for steel bridges than for con-
of ambient temperatures.
in both simple span and continuous bridges and assume that movements
provisions are made for dead and live loads; 3) the primary considera-
Since 1960 much work has been done in the area of thermal strains.
Several bridges have been observed by Zuk (76), each during a one-year
bution which would result from a set of given conditions. Daily and
3
concrete bridge, the concrete deck heats up from the top do\vo through
the slab due to solar radiation, while the steel is shaded most of the
stress, or both strain and stress, may result (41). Thermal strain
inward, some point on the bridge deck will not move. This point--
other portions of the deck are defined, and a better idea can be had
bearings. The horizontal forces» which are transmitted to the piers and
abutments due to the movement of the bridge deck» can then be more
This paper seeks to acquaint the reader with the state of the art
the thermal effects on bridges are reviewed and grouped for continuity
AIR TEMPERATURE
There are two basic temperature cycles (9). The daily cycle
usually begins with a low temperature being attained just before sun-
hours before sunset, and then air temperatures drop more rapidly to a
low reached prior to sunrise the next morning. The basic daily temper-
ature cycle may be altered by the presence of clouds shading the area
drop of temperature. New air masses moving into the locality from a
cooler or warmer region may also mask the usual daily temperature
cycle. The yearly temperature cycle results from the changes in posi-
tion and distance of the earth relative to the sun. Both of these
parts of the structure while the yearly cycle induces the greatest
(25) used meteorological records showing the absolute and the average
7
priate to any area of Britain, and that the minimum temperatures listed
were too high. Other British researchers (18,58) have also used
BRIDGE TEMPERATURE
temperature as that of the air (76). The upper elements and the
solar radiation received, the wind, and the amount and type of pre-
cipitation (9). The top of the slab is warmer than the bottom of the
bridge when the sun shines on the exposed deck. The top will cool
faster than the girders when a rain or snow storm first begins. A
uniform temperature can exist just before sunrise when the air tempera-
ture has remained nearly constant for several hours. Thus, a variety
bridge.
the temperature of the steel girder. The first case involved a uniform
temperature throughout the slab. Cases two and three had a non-linear
temperature variation in the slab, one with the top of the slab warmer
than the bottom and one with it cooler. Maher (41) assumed a linear
box sections.
top slab of the box section. The sides and bottom of the box section
composite with a concrete slab. They extended Zuk's earlier work with
for each type of member. The shear and the moment at the interface
between the beam and slab and the force in the tendon were all deter-
ea = air temperature, in °C. Wroth (67) used the interior air tempera-
England: the Medway Bridge (70% concrete beam and slab, 30% variable
depth concrete box); the Hammersmith Flyover (concrete box spine beam) ;
out the depth and width of each bridge. All strain measurements were
attempt to separate the thermal strains from those which might result
for the 66-ft composite bridge (76). Then a one-inch thick coating
the bridge and temperatures were recorded for an additional nine months.
Typical temperature distributions were shown for both cases and the
deck. This was accomplished by the heat flow up from the insulated
beams into the slab which heated up the slab a few degrees warmer than
11
it would have been without the insulation. One problem noted was the
period for the 36-ft reinforced-concrete bridge, and some typical values
including the thermal contours through the beam and slab were presented
ior beams exposed to the sun and the shaded interior beams.
for two conditions. For a normal concrete deck in the Middle Atlantic
pyrheliome ter.
12
where A is the factor indicating the phase lag between the maximum
appropriate for the summer and a value of one-half for the winter.
were compared with the results of field tests conducted on the 66-ft
differential between the top and bottom of the bridge checked very
the thickness of the slab with the calculated values showed fair agree-
Emerson (25) found that in England the maximum range of mean bridge
below the minimum temperature to about 1.5 times the maximum tempera-
of -6°F at the bottom of the steel girder with an equal air temperature.
cent greater than those which occurred before the insulation was applied.
This increase was due to the slower transfer of heat from the air to
were observed at the top surface of the two-inch thick asphalt surfac-
ing with an air temperature of 93°F. The concrete slab reached its
highest temperature about two hours later (4:30 p.m.) with the top
temperatures were higher than the air temperatures at the top and
those stresses caused by live and dead loads. However, when the top
of the slab is cooler than the bottom, the stresses induced by the
as the two materials try to match the movements of each other (9).
between summer and winter which will cause general expansion or con-
stresses induced when the materials are restrained from movement. The
first method is commonly used for simple span bridges and the second
is often used for continuous span bridges, but either method can be
used with either type structure. In any case the designer needs to
know what thermal stresses and/or movements will occur. Several types
of bridges have been studied to determine the movements that occur with
varying temperatures.
temperature and humidity movements; and 3.3-in. for creep and shrink-
residual creep and shrinkage. The air temperature inside the box-
The movements measured at one of the pier columns closely followed the
ship between these movements and the interior air temperature was
case, the five-day mean bridge temperature (calculated from the move-
ing station were quite comparable. Thus the probable range of five-day
mean ambient temperature. The impetus for this investigation carne from
the trend toward the construction of more bridges with longer spans
(100 to 200 ft) which increased the problems with joints and bearings
movements was discussed. However, the authors did not, at this stage
weight structures with longer spans. The longer spans lead to propor-
were 1/2 in. and 3/4 in./IOO ft of span for concrete and steel bridges,
respectively.
linear thermal gradient patterns. In the analysis, the beam and slab
at the interface and equal curvature of the two parts at the interface.
The stresses (both axial and lateral) can be determined in any part
SLAB:
f
3
xs(y) = F/(2ap) + 3ys (Fa - Q)/(2a p) (5)
in.
BEAM:
Y
b
= distance measured from the mid-thickness of the slab (+ down),
in in. Some values calculated with these equations, such as longitudin-
Liu and Zuk (40) extended Zuk's earlier work (72) with composite
were calculated using the more complex equations developed for four
the slab in a direction transverse to the beams varied from 1000 psi in
that direction.
out the depth of the bridge and the difference between the coefficients
actual movements with the design values t it was found that several
were calculated for concrete and steel t respectivelYt for the struc-
of box girder structures was less than that of other concrete structures
due to the insulating effects of the air trapped inside the box section.
bottom of the girder to the temperature difference between the top and
21
f = 2500T /h (8)
b s
in which f
b
= the thermal stress in the bottom flange of the girder
(+ equals tension), in psi; T the temperature difference between
s
the top and bottom of the slab (top temperature less the bottom temper-
tion of air temperatures and bridge movements. The data showed that
end movements of the bridges, but that other factors such as creep,
of expansion for the bridges, top and bottom, for both winter and summer
and the yearly extremes was also presented. It was noted that the
values are widely scattered. No specific reason for this was given
change due to moisture, and dead and live loads. As a guide for
The equations developed for deflection of the slab and beam were
the thermal loads may vary throughout the thickness of the beam and
approximate both the mathematical model and the highway bridge slab
resultant deflections were too small, even for large thermal gradients.
The model was abandoned and replaced by an 8-ft by 6-ft by 3-in. thick
concrete slab with two layers of welded steel fabric serving as rein-
forcement--one near the top surface and one near the bottom. The
ior of the heated slab. The slab arched up when the heat was first
applied due to the top heating up more quickly than the bottom of
and in some cases actually reversed t even though the temperature dif-
ference between the top and the bottom of the slab remained nearly the
the top of the slab was warmer than the bottom. The thermal stresses
calculated from strains measured in the model were quite high but not
for two days in August and one 22-hour period during a storm in
shown for the two days in August. Creep would not be as significant
calculated from the mathematical model. This reportedly was due pri-
marily to the deviation of the bridge from the theoretical model and the
the top of the slab. These stresses would be offset by the compressive
stresses ordinarily found in the top of the slab due to dead and live
950 psi. These stresses would be additive with those of dead and live
loads.
CURRENT SPECIFICATIONS
the code are especially pertinent to this study (4, p. 25, 56).
1.2.15--THERMAL FORCES
Metal Structures
Temperature Temperature
Concrete Structures rise fall
Moderate climate 30 F. 40 F.
Cold climate ......• 35 F. 45 F.
l.5.4--EXPANSION
bridges, perhaps because the concrete does not react as quickly to the
25
heating and cooling effects of weather changes. The larger mass and
from the 1963 standards which provided for a 58.5°F temperature variance.
disclosed that only the 1965 (now 1969) AASHO specifications make a
Austria, Sweden, and Japan are the only other countries with a
thermal stress provision in their codes, and these codes are all
essentially based on the German code. The United States has no
direct provision, although Section 1.2.15 of the 1961 AASHO
Standard Specifications for Highway Bridges states: "Provision
shall be made for stresses or movements resulting from variations
in temperature."
27
EXPANSION JOINTS
joints if they are to allow for the smooth passage of vehicles, while
not restricting the free movement of the bridge. The joints are in-
which span and cover the gap; 4) elaborate sliding plate joints which
have additional devices to help span the gap resulting from larger
Agnew, and Palmer (30) have developed joints for large (up to 73 in.)
Berks has developed four joints for various ranges of movement based
and pulled down with considerable force (12 tons/ft run) throughout
their length. The French recommended that the bolts holding down the
increase rapidly as they become more complex. For example, using the
pansion joints are twice as expensive, while finger expansion joints are
anchoring the slab to flexible stub abutments and piers. This has
become generally accepted and is being used for both steel and con-
ever, opinions vary as to how to determine and provide for the stresses
necessary.
29
JOINT SEALANTS
deicing chemicals.
Dreher (22). Rather than merely using the slot as a mold for the joint
sealer, his objective was to form a bridge between the two elements
of the deck. Hinged levers, arched springs, and other more elaborate
and thus would not rely on a bond with the sides of the gap to main-
Cook (21) found that the properties of high bond strength, ex-
sealants made them very desirable for use. Additional study is re-
to be subjected.
by Kozlov (37).
prefabricated at the factory and placed in the deck before the con-
being extruded from the joint opening; 5) survive the wear and impact
from poor joint sealing were shown (61,62). Such problems as salt
BRIDGE BEARINGS
"a seat for a girder or truss that provides the necessary reaction
bearings and rockers were later developments, but the need to provide
for movement and rotation in both the longi tudinal and transverse
directions led them to become bulky and unwie1ding. Nevin (46) suggest-
small loads and movements, and the sliding plate technique for
less could be expected with Teflon coated plates under suitable con-
They can be designed to resist large vertical loads with little deform-
(5). These bearings have proven satisfactory and economical for use
with concrete and steel bridges (28). Their low cost, freedom from
maintenance, and low profile has led to their wide use as bridge bear-
ficantly lower (15%) than for conventional rigid bearings (73). The
additional savings over the cost difference between them and metal
be made for impact. McKeel and Kinnier (43) found from field tests
that little advantage of reduced stresses would be gained from the use
these results with both the theoretical study of Zuk (73) and the ex-
extremely well over the past ten years [1957-1967] and all installations
structure. "
35
FUTURE NEEDS
will allow the design engineer to evaluate both the benefits and
SUMMARY
top and bottom of the bridge based on weather data. Test results and
occur only under special conditions, such as during the winter when
clouds screen the bridge deck from the sun's effects. The use of a
the warmer parts of the bridge expand more than the cooler parts. A
ture distribution through the concrete slab, while the steel beam is
distribution through the top slab is often assumed for concrete box
sections; the sides and bottom of the box have been found to have
complex. Use of flexible stub abutments and piers eliminates these costs.
APPENDIX I.-BIBLIOGRAPHY
13. Black, W., and Adams, H. C., "Survey of Expansion Joints and
Bearings in Britain," Int. Assn. for Bridge and Structural Eng.,
7th Congress, Rio de Janeiro, Preliminary Publ., Vol. 1, 1964,
pp. 1029-1039.
23. Ekberg, C. E., Jr., and Emanuel, J. H., "Current Design Practice
for Bridge Bearing and Expansion Devices," Final Report, Project
547-S, Iowa State University, Engineering Research Institute,
Aug., 1967.
27. Erickson, E. L., and Van Eenam, N., "Application and Development
of AASHO Specifications to Bridge Design," Journal of the Structur-
al Division, ASCE, Vol. 83, No. ST4, Proc. Paper 1320, July,
1957, pp. 1-38.
29. Fukuyama, T., "Effect of Supporting Piles on the Stress and Deforma-
tion of Reinforced Concrete Rigid Frames," Paper SP 23-44, Concrete
Bridge Design, American Concrete Institute, Detroit, 1969, pp. 783-
801.
30. Gortz, W., Agnew, F. S., and Palmer, M. F., "Bridge Deck Expansion
Joints," Consulting Engineer, London, Vol. 31, No.5, May, 1967,
pp. 76-78, 81, 83.
32. Hendry, A. W., and Page, J. K., "Thermal Movements and Stresses
in Concrete Slabs in Relation to Tropical Conditions," RILEM
International Symposium on Concrete and Reinforced Concrete,
Part 2, July, 1960, PP. 1-26.
38. Koz1ov, G. S., and Desai, D., "Preformed E1astomeric Bridge Joint
Sealers: Thermal Characteristics of Bridge End Movements,"
Highway Research Record, No. 302, 1970, pp. 38-48.
43. McKeel, W. T., Jr., and Kinnier, H. L., "Dynamic Stress Study of
Composite-Span Bridge With Conventional and Elastomeric Bearings,"
Highway Research Record, No. 354, 1971, pp. 13-26.
45. Naruoka, M., Hirai, I., and Yamaguti, T., "The Measurement of
the Temperature of the Interior of the Reinforced Concrete Slab
of the Shigita Bridge and Presumption of Thermal Stress,"
Symposium on the Stress Measurements for Bridge and Structures,
Proceedings, Japanese Society for the Promotion of Science,
Tokyo, Japan, 1957, pp. 106-115.
49. . .
Pare , R L , "Neoprene Elastomer Bearings--Ten Years Experience
Proves Their Importance," Civil Engineering, ASCE, Vol. 37,
No. II, Nov., 1967, pp. 37-39.
42
79. Zuk, W., "End Movement Studies of Various Type Highway Bridges,"
Virginia Highway Research Council, Aug., 1968.
80. Zuk, W., "End Movement Studies of Various Type Highway Bridges,"
Highway Research Record, No. 295, 1969, pp. 1-4.
45
APPENDIX II.-NOTATION
f
b
= thermal stress in bottom flange of the girder, in psi;
f
Xb
longitudinal stress in beam (+f = tension), in psi;
f
xs = longitudinal stress in slab, in psi;
f
zs = transverse stress in slab, in psi;
h total depth of bridge, in in. ;
4
I moment of inertia, in in. ;
m = Poisson's ratio;
p = slab width, in in. ;
T
a = average daily temperature, in of;
T = maximum surface temperature, in of;
m
T = daily range in temperature, in <IF;
r
T = temperature difference between the top and bottom of the slab
s
(top temperature less the bottom temperature), in of;
46
t time, in hours;
in in.;
A = lag factor.
47
VITA