Box Girder Bridge Design—State of the Art

C. p. HEINS

The information that will be presented herein will pertain where

to straight and curved steel composite box girder bridges
AT,AB total area of the top and bottom flange, re-
of moderate span length (50-250 ft) that are utilized for
spectively, in.^
highway interchanges. Although the general theories are
L span length, ft (90 < L < 200)
applicable to larger structures, the design formulas that
d girder depth, in.
have been developed and will be given herein are only
suitable for the conventional highway bridge.

BRIDGE TYPE
Details—During the past two years, the ASCE Task
Committee on Horizontally Curved Steel Box Girder
Bridges has conducted a comprehensive survey^ on the
details of box girder bridges (straight and curved). This
survey has shown that the number of box girder bridges
being built has increased dramatically since 1961, as shown
in Fig. 1.
The collected details (geometry) of the 82 reported
bridges has been reduced to those shown in Tables 1-3,
with the ratios of these dimensions given in Tables 4-6. The
various parameters listed in these tables are shown in Fig.
2. These bridges represent typical steel-composite struc-
tures, and the data given therein can be used for prelimi-
nary design. The bridges generally have internal cross-
I I I I I I I I
bracing and top lateral bracing with external diaphragms 61 63 65 67 69 71 73 75
placed only at the support piers.
Section Geometry—In the design of any complex struc-
ture in which the section changes and the forces are not Fig. 1. Frequency of box girder bridge built vs. year built
readily computed, it is useful to have data or empirical
relationships to select plate and girder sizes. Examination
of the data presented in Tables 1-3 has resulted in the
following general equations:
Single Span:

(1)

AB = 12.9a' (2)

C. p. Heins is Professor of Civil Engineering, University of

Maryland, College park, Md.
Ed. Note: This paper was originally presented at the AISC Na-
tional Engineering Conference, Los Angeles, Calif, in May
1978. Fig. 2. Box girder dimensions

126
ENGINEERING JOURNAL / AMERICAN INSTITUTE OF STEEL CONSTRUCTION
 Table 1. Single-Span Box Bridge Dimensions

Brace
L R d B bf d Spac'g
(ft) NB (ft)„ (ft) (in.) (in.) (in.) (in.) (in.) (in.) (deg) (ft)

90 1 6.6 300 36.6 50 % 7. 12 78 0 a

109.6 2 12-16 760 52 72 178 78 18 1 0 11
118.5 4 43-48 1773.8 48 60 % % 18 1 0 a
560.5
140 4 55.6 600 120 56 2 % 24 1-274 13.6 25
177 4 55.6 2000 120 56 278 % 24 2-278 13.6 25
208 5 78.8 0 77 61 2 % 24 2 13.2 23-26

^Not known.

In addition to developing Eqs. (1) and (2), the following 3. Width of box varied between 50 in. and 72 in.
general trends were noted: 4. Maximum thickness of bottom flange was 2 in.
5. Width of top flange varied as Wp = 0.12(L - 100)
1. Web thickness varied from V2-in. to %.-in.
+ 12 with a minimum of 12 in.
2. The ratio of roadway width in feet, WR, to number 6. Maximum thickness of top flange was 2% in.
of boxes, NB, varied as 6.5 < WR/NB < 16.0, with 7. Minimum ratio d/L was 1/30, with an average of
an average of 12.0. 1/23.

Table 2. Two-Span Continuous Box Bridge Dimensions

Brace
L WR R d B ff Spac'g
(in.) (in.)
tb
(in.)
^w
(in.)
V
(in.) (in.)
d ^y
(ksi)
(ft) NB (ft) (ft) (deg) (ft)
100 3 44 oo 48 72 + 1 % % + 12 + % a a 36
100 -17
108 2 30 162.5 48 57
-
+
%
7.6 \ 14
-n% 26 15 50
108 - 7.
111 2 42 2885 42 114 + % % + 12 + i'A 8 15 50
111 - 78 -18 -2V4
120 3 44 oo 58 68 + 78 % + 14 + 78 a a 36
120 - Vs -19 -1%
120 2 42 1637 56 108 + 7l6 + 7,. +21 +1 9 16-8 50
120 - 78 - % -24 -l'/4
145 3 43 oo 60 56 + "A6 7,. 14 + '7.. 0 a 50
145 - \ -l'/3
145 3 43 oo 60 56 + 78 7s 14 + 78 0 a 50
145 - %e - "L
174 2 38 2272 71 101 + % 7» 18 + 1V4 9 24 50
174 - % 18 - 2 7.
185 2 55.6 oo 120 56 + 1 7. + 74 +24 +1 0 a 36
185 - 2 7. -1 -66 -I'/s
194 3 42 2884 63 86 + 1 % 7s 24 2V4 0 12 36
214 - 2 74
217 2 55.6 2000 120 56 + 2 % + % +24 + 174 0 a 36
217 -2 78 -2 -66 -17s
220 2 55.6 2000 120 56 + 2 78 + 74 +24 + l'7,e 13.6 a 36
220 -2 -66
[ - 3 78
-n
Note: + indicates positive moment region. — indicates negative moment region.
^Not known.

127
FOURTH QUARTER / 1978
Two Span:
The total plate areas of a two span bridge, shown in Fig.
3, are computed as: WWW
r r Twyrr-

100 ' L < 2 2 0 ' A,

AB^ = ^ (0.00153L2 - 0.223L + 13) (3)

FyT Fig. 3. Two-span box girder bridge—flange area locations

AB-=IA7AB+ (4)

0.64 ^B+ (5)

P.-
= 1.6^fi+ (6)
F + Py = yield point of material, ksi, at section
^y being examined
where
L = span length, ft (100 < L < 220)
AB^,AB~ ~ total bottom and top flange areas K = NBFyd/{WR X 600), where Fy is at
AT^, AT~ (in.2) in the positive (+) and negative AB^
(—) moment regions, as shown in WR = roadway width, ft
Fig. 3 NB = number of boxes

Table 3. Three-Span Continuous Box Bridge Dimensions

Brace
L R d B d Spac'g
^R
(in.) (in.)
tb
(in.)
^vy
(in.)
V
(in.) (in.)
'f
(deg) n
^y
(ft) NQ (ft) (ft) (ft) (M
90 2 44 716 41 94 \ 7s + 14 + 74 0 18^ 1.33 +36
120 -22 -1 24 -50
90
93 2 38 2845 42 114 + % % + 12 + 7s 9.5 15 1.33 50
124 - \ -16 -1
93
102 2 25 650 51 557, + 7s - \ a a 14 10.3 1.0 36
102 - 7s
102 + %
104 2 38 5730 61 108 + \ Is + 16 +1 9.5 15 1.54 50
160 -1 -16 -2
104
100 2 43 1310 54 1067, + \ + % +24 + IV4 14 a 1.08 36
132 -1 - % -30 -174
122
116 2 33 oo 63 8174 + 7s % 14 + 74 a a 1.13 a
131 - 74 -IV.
116
166 4 68.5 1000 55 78 + 1 % 7s +20 + 178 8 20.6 1.24 50
206 - 2 74 -30 -3
166 + 1 +20 + 178
173 3 40.5 760 60 92 + 1 V4 + 78 +20 + 174 0 a 1.14 +36
196 - 1 V4 - % -24 -274 -50
172 +1 +20 + 1
174 2 55.6 2000 120 56 + 1 + 74 +24 +1 13.6 a 1.62 36
282 - 3 78 -2 -69 -17s
125 + 3 78
100 2 Var. 874 120 48 + 78 + % 16 +1 0 10 1.0 +36
100 - %e - 7,. -2 -44
100

Note: + indicates positive moment region. — indicates negative moment region.

^Not known.
^At end.

128
ENGINEERING JOURNAL / AMERICAN INSTITUTE OF STEEL CONSTRUCTION
The general data indicates that: Three Span:
1. Web thickness, tyj, varied from %-in. to •y4-in. The total plate areas of a three span bridge shown in Fig.
2. WR/NB varied as 14 < WR/NB < 28, with an 4 are computed as:
average of 20. n
3. Width of box B, in inches, varied approximately as AT^ = (Li - 73) (7)
6AK
B = 2.5[{WR/NB) - 11], with WR in inches, and
^ > 55 in.
4. Width of top flange, bj, varied from 12 in. to 24 in., AB+ = f^iL,-52) (8)
in accordance with 6/+ = 0.12(L - 100) + 12.
5. Top flange thickness tj was less than 2V2 in. in all
cases. AT- = (Li - 100) (9)
2.6K
6. {d/L)^^n = 1/30, with {d/L)a,g = 1/25.
AB- = -TT (0.964L2 - 1.65(10-VZ^22) - 70) (10)
An
Table 4. Single-Span Box Bridge Geometric Ratios AT+ = 0.95 AT- - 0.011(ylr~)2 - ^A/K (11)
L
(ft) d/L L/R L/tt d/B B/tb VA/ AB- (L2 - 48) (12)
\0K
90 0.0339 0.300 2160 0.732 100 13.7
109.6 0.0395 0.144 956.4 0.722 52.4 12.0
118.5 0.0338 0.211 2844 0.800 120 18.0
140 0.0714 0.233 840 2.14 28 10.7
177 0.0565 0.0885 679.2 2.14 17.9 9.14
208 0.0308 0 1248 1.26 30.5 12.0
Table 6. Three-Span Continuous Box Bridge
Geometric Ratios
L d/L L/R Ljtb d/B BItb bfltf
Table 5. Two-Span Continuous Box Bridge 90 0.0285 0.168 1644 0.436 107.4 + 18.7
Geometric Ratios 120 -14.7
90
L
93 0.0283 0.0436 2976 0.368 + 228 + 13.7
(ft) djL LjR LItb djB B/tb bfltf
124 -9.85
100 0.040 0 2400 0.667 + 144 + 24 93
100 -96 -11.3 102 0.0417 0.157 2448 0.919 +63.4 a
108 0.0370 0.665 4152 0.842 + 182 + 22.4 102 -63.4
108 -114 -22.4 102 + 111
111 0.0315 0.0385 2664 0.368 +228 +8.0 104 0.0318 0.0279 2556 0.565 + 144 + 16.0
111 -130 -8.0 160 -108 -8.0
120 0.0403 0 2304 0.853 + 109 +22.4 104
120 -77.7 -12.7 100 0.0341 0.101 2114 0.507 + 142 + 19.2
120 0.0389 0.0733 3288 0.519 +247 + 21.0 132 -107 -17.1
120 -173 -19.2 122
145 ' 0.0345 0 2532 1.07 +81.5 + 14.9 116 0.0401 0 4188 0.771 + 218 + 18.7
145 -74.7 -9.3 131 -109 -9.33
116
145 0.0345 0 2784 1.07 +89.6 +22.4
145 -81.5 -17.2 166 0.0223 0.206 2472 0.705 +62.4 + 12.3
206 -34.7 -10.0
174 0.0340 0.0766 4176 0.703 + 202 + 14.4
166 + 78.0 + 12.3
174 -135 -7.2
173 0.0253 0.261 2376 0.652 + 73.6 + 11.4
185 0.0573 0 1476 2.14 +37.3 +24.0
198 -73.6 -10.7
185 -24.9 -Closed
172 +92.0 +20.0
194 0.0263 0.0707 1596 0.733 +57.3 + 10.7 174 0.0355 0.141 1082 2.14 +56.0 +24.0
214 -38.2 -10.7 282 -17.9 -36.8
217 0.0461 0.109 1157 2.14 +24.9 + 13.7 125 + 17.9
217 -19.5 -Closed 100 0.100 0.114 2136 2.5 + 76.8 + 16.0
220 0.0454 0.110 1006 2.14 +21.3 + 12.4 100 -85.3 -8.0
220 -17.9 -Closed 100

129
FOURTH QUARTER / 1978
 Ax ^AT / *T

-cp- -^^ ^c
"isr 1

Ag Ag Ag 1 ' t
A- I

' 1 ' t
Fig. 4. Three-span box girder bridge—flange area locations 1 ' 1
1 B 1
Fig. 5. Actual braced section
where
9 0 < L i < 180 (ft)
1 0 0 < L 2 <290(ft)
n = L2/L\
NBiFy)d
^ (WRX 600)
Fy = Specified yield point of material at section being
examined
A study of the data indicates that:
1. Web thickness, t^, varies from ^/g-in. to Vg-ii^-
2. WR/NB varies as 12.5 < WR/NB < 22, with an
average of 18.0.
3. The width of the box varies as 5 = 7A3[{WR/NB)
— 5.0], with Bmin - 55 in. and B^ax = 114 in.
4. The top flanges had geometry of '^U — ^/ — ^ and
12 < fe/< 30.
Fig. 6. Modeled braced section
5. {d/L2)rmn = 1 / 4 0 w i t h {d/L2)avg = 1/30.5.

LOAD DISTRIBUTION
Curved Bridges—If the bridge system is composed of
Straight Bridges—the AASHTO specification^ provides
curved box girders, then resistance of these elements to the
design criteria for evaluation of the induced bending mo-
applied live load is accomplished by interaction of torsion
ment in straight composite multi-box girder bridges of
and bending. This interaction creates a highly indetermi-
moderate length. This provision, based on research work
nate situation and, in general, requires utilization of a
by Mattock^ and the box girder geometry listed in Tables
computer program."^ In order to minimize the need for a
7 and 8 and Figs. 5-8, is Sect. 1.7.103:
computer solution, the response of the typical composite
"The live load bending moment for each box girder sections, shown in Figs. 7 and 8 and Tables 9 and 10, have
shall be determined by applying to the girder the frac- been studied when subjected to live loading with the radius
tion WL of a wheel load (front and rear) according to 200 < R < 10,000 ft. The resulting maximum moments
the following:" were then related to those in the straight system as a func-
tion of Wi^, as given by Eq. (13). The resulting
0.85 equation is:
WL = 0.1 + \.7R + (13)
WL, = (1440^2 + 4 8 ^ + i ) j ^ ^ (14)
where
where
R = M, 1.5 >R >0.5 ^Lc ~ curved girder load distribution for moment
number of box girders ' Wt = straight girder load distribution
N^ = Wc/\2, reduced to nearest whole number X = \/R
Wc = roadway width between curbs, ft R = center line radius of bridge system, ft

130

ENGINEERING JOURNAL / AMERICAN INSTITUTE OF STEEL CONSTRUCTION

 i
f .
UJ4 Concrete Slab

TT
T '
5

7T, Lateral Bracing

(Equivalent)
T '__#„

13-

'
t
B'

Fig. 8. Modeled composite section

Fig. 7. Actual composite section

Dead Load
Bending—The dead load bending moment^'^ is evaluated
ESTIMATED CURVED GIRDER FORCES from:
T h e effect that curvature has on the internal bending mo- M, = {KD)(wLVS) (15)
ment of a box section, when subjected to truck live load, can
be determined readily by using Eqs. (13) and (14). T h e where
determination of dead load moment for a curved simple
span box can also be readily determined by applying the 1
KD = \ +
following. 10(i?/L)2

Table 7. Dimensions of Braced Sections

Dimensions (in.)
Bridge Span
Type (ft) A B C D n T2 TS 74 T5

2L2G 108 96 12 19.44 0.625 0.375 0.5 0.0878 0.217

3L 3G 50 98 86 12 19.47 0.5625 0.375 0.5 0.0980 0.224
4L4G 92 80 12 19.50 0.5 0.375 0.5 0.105 0.222

2L2G 110 96 14 48.125 1.0 0.5 0.75 0.0878 0.331

3L 3G 100 99 86 13 48.125 1.0 0.4375 0.75 0.0980 0.348
4L4G 92 80 12 48.125 1.0 0.4375 0.75 0.105 0.352

2L2G 110 96 14 74.75 1.5 0.625 1.0 0.0878 0.458

3L 3G 150 99 86 13 74.75 1.5 0.5625 1.0 0.0980 0.479
4L4G 92 80 12 74.75 1.5 0.5625 1.0 0.105 0.483

Table 8. Dimensions of Composite Sections

Dimensions (in.)
Bridge Span
Type (ft) A' B' C D' E' F' TV T2' r3' r4' Tb'

2L 2G 192 ..96 12 19.44 25.75 42.0 0.625 0.375 0.5 0.0878 0.947
3L 3G 50 172 86 12 19.47 25.75 37.0 0.5625 0.375 0.5 0.0980 0.955
4L4G 160 80 12 19.50 25.75 34.0 0.5 0.375 0.5 0.105 0.958

2L 2G 192 96 14 48.125 54.625 41.0 1.0 0.5 0.75 0.0878 1.019

3L 3G 100 172 86 13 48.125 54.625 36.5 1.0 0.4375 0.75 0.0980 1.030
4L4G 160 80 12 48.125 54.625 34.0 1.0 0.4375 0.75 0.105 1.033

2L 2G 192 96 14 74.75 81.5 41.0 1.5 0.625 1.0 0.0878 1.092

3L 3G 150 172 86 13 74.75 81.5 36.5 1.5 0.5625 1.0 0.0980 1.106
4L4G 160 80 12 74.75 81.5 34.0 1.5 0.5625 1.0 0.105 1.108

131
FOURTH QUARTER / 1978
 L = span length, ft
R = radius to center line of box, ft
w = dead load per unit of length •
Torsion—The induced torsional dead load force can sim-
ilarly be determined by:
Tc = (0.75^)(z^L) (16)
where e = center line offset of the curved box.
Bending Normal Stres

Live Load
Fig. 9, Bending normal stress
Torsion—The live load bending effect is found by Eqs. (13)
and (14). The torsional force is estimated as follows:
Tc = 72NT{0.9Se + X) (17)
where 72 represents the gross weight of AASHTO truck,
and : YQ

A^T^ = number of trucks on box

X = eccentricity from center line of box to resultant
of laterally positioned trucks
e = center line offset of curved box Fig. 10. Bending shearing stress

T a b l e 9. Section Properties of M o d e l e d B r a c e d Sections

Bridge Span Y K'T' BENDING AND TORSIONAL STRESSES

Type (ft) (in.) (in/) (in.^) (in.^) (in/) Straight Bridges—A straight box girder, when subjected
2L2G 6.94 6,550 18,900 3,220,000 1,620,000 to live loads, will develop bending stresses. The inducement
3L 3G 50 7.16 6,090 17,000 2,200,000 1,230,000 of torsional stresses may occur when the loading is eccentric
4L4G 7.21 5,720 15,600 1,710,000 1,020,000 to the shear center. However, in general, such effects are
18.6 67,400 140,000 9,450,000 18,400,000
neglected. Thus the stresses that are considered are:
2L2G
3L 3G 100 18.9 61,700 118,000 2,940,000 13,500,000
fb = M/S (18)
4L4G 19.1 58,300 107,000 1,720,000 11,100,000
and
2L2G 30.3 236,000 378,000 3,710,000 68,000,000
3L3G 150 30.7 216,000 311,000 216,000 49,500,000 T= Vd/It (19)
4L4G 31.0 205,000 280,000 528,000 41,300,000
where
M induced bending moment in the box
Table 10. Section Properties of Modeled S section modulus
Composite Sections V induced bending shears in the box
Bridge Span Y Q statical moment
Ix
Type (ft) (in.) (in/) (in/) (in/) (in/) I moment of inertia
t plate thickness
2L 2G 19.8 27,200 56,800 3,640,000 8,280,000
3L 3G 50 19.8 24,700 49,100 2,170,000 6,140,000 These typical stresses are shown in Figs. 9 and 10.
4L 4G 19.7 23,100 44,600 1,540,000 5,050,000

2L 2G 37.8 178,000 250,000 15,800,000 67,800,000 Curved Bridges—As in the case of straight bridges, curved
3L 3G 100 37.9 159,000 197,000 16,400,000 48,400,000 bridges will develop bending stresses. However, due to the
4L4G 37.8 150,000 176,000 15,400,000 40,100,000 curvature of the bridge, torsional forces will also develop;
the magnitude of these will depend on the cross-sectional
2L 2G 52.1 511,000 551,000 98,700,000 228,000,000
3L 3G 150 52.3 460,000 433,000 101,000,000 164,000,000
geometry, span length, and radius.
4L4G 52.1 433,000 385,000 92,400,000 137,000,000
Pure Torsion—Any section, open or closed, when subjected
to a torsional loading, will resist (in part) the applied torque
Krp= pure torsional constant by pure torsion given by:
/yy = warping constant
TpT = GKT(t>' (20)

132
ENGINEERING JOURNAL / AMERICAN INSTITUTE OF STEEL CONSTRUCTION
where A comparison of the stiffness KT for typical open/closed
box sections^ indicates that KT^iosed = 10"^ >< ^Topen T h u s ,
ATj = torsion constant if one can structurally close a box section, tremendous
G = shear modulus torsional stiffness can be achieved.
(f)^ = rate of change of rotation per unit length
T h e induced shearing stresses are given by
Equation (19) is probably more familiar in the following
form: TpT — (23)
KT

T^GKTJ or 0 = TL/GKT where t = plate thickness of any element.

as given in texts on strength of materials. Warping Torsion—When a thin-walled section is subjected

T h e evaluation of the torsion constant KT is dependent to torsional loading, the elements do not retain their shape
on whether the box section is open or completely closed, and thus warp. This warping will induce normal stresses
where ''closed" would represent a composite or braced and shearing stress, which can be substantially significant
section. T h e determination of KT is given^ by the fol- in open sections. T h e general warping normal stress is
lowing: (24)
Open box: and warping shear is given by

KT = -[Link]^ (21)
Trn = •
t
where where
h = larger plate dimension Bi = EI^(j)^' = warping moment or bimoment
t = smaller plate dimension for same element W„ = normalized warping function
Closed box: I^ = warping constant
<j> = total angle of rotation, radians
ds warping statical moment
KT = 4AoVf (22)
These three induced stresses, for a closed box, are shown
where in Figs. 11-13.
AQ = enclosed area of box section, luR to center line Limiting Effects—The determination of all stresses in a
of elements curved box is difficult, due to bending and torsion inter-
ds = length of a given element action. T h u s , it would be desirable to determine if it is
t = corresponding thickness of that element necessary to evaluate both warping and pure torsional
stresses in a box girder bridge.
A recent study^ has indicated that for a central angle d
between 0 and 0.5 and)// > 10 + 40^, warping can be dis-
regarded; for d between 0.5 and 1.0 and \p > 30, warping
is negligible, where xp = L[GKT/EI^]^^^. A study of many
curved bridge systems and their respective ^p parameters,
as shown in Fig. 14, shows that for single closed box units,
\t > 30; thus warping can be disregarded, providing the
Fig. 11. Pure shearing stress, TPT section is closed.
Pure torsional stress can be disregarded when \p <
0.4.

Fig. 12. Normal warping, f^ Fig. 13. Warping shear, r^

133
FOURTH QUARTER / 1978
 Normal Warping f„

Transverse Bending m^.

Figure 14 Fig. 16. Distortional stresses

bending moment, as shown in Fig. 16. In order to inhibit

such a response, internal diaphragms can be used^ and thus
minimize the stress development.
The required spacing and size of such bracing will be
discussed in the next section.

BRACING
Fig. 15. Shear distortion Top Lateral Bracing—As discussed in the section on
Bending and Torsional Stresses, the torsional stiffness KT
can be substantially increased if a box section is completely
closed. Examination of Figs. 17 and 18 shows a box section
with top lateral bracing. This bracing can be converted^ into
DISTORTIONAL STRESSES an equivalent thickness by the following equation:
In the development of the general torsional equations, it is
E2Aa
assumed that the section retains its shape during defor- ^ea ^^ ' (25)
mation and final stress evaluation. However, v^hen a box
^eq
G b cos^« sma
section is subjected to torsional loadings, its cross section A study of various curved box elements, subjected to
does not retain its shape,^'^^ ^g shown in Fig. 15. Such a induced bending and warping normal stresses, has indicated
response will create additional stress in the section and will that for 1 < b/d < 3 (width of box to depth of box) and a
include a normal stress, a shearing stress, and a corner stress ratio <10%, t^q < 0.050 in., as shown in Fig. 19.

134
ENGINEERING JOURNAL / AMERICAN INSTITUTE OF STEEL CONSTRUCTION
 -f- - -t:

Fig. 18. Top lateral bracing

Diaphragm Bracing—As indicated previously, the min-

imization of normal stresses due to distortion can be
achieved by placement of internal cross-diaphragms, shown
in Fig. 20. A study by Heins and Olenik^, using the typical
sections shown in Tables 7 and 8, has resulted in the fol-
Fig. 17. Top lateral bracing lowing general equation:
fd _ (lOL - 350) S^
(27)
fb R L2
Thus, the bracing area required would be:
where
^^(in.2) > 0.036
fd = induced dead load distortional normal stress
where b = bottom flange width, in. fb = induced dead load bending normal stress
In addition to minimizing the warping stress, the section L = span length, ft
becomes closed and thus yp > \0, which also satisfies the R = radius, ft
warping criteria. S = diaphragm spacing, ft

\\\ ' \ b/d = 2..

\ \ \ E 2A,
\ \ t = - d (.3535)
%' yb/d=l. ^ -q C b
\ \ ' R = 400 ft.
b/d = . 5 \ \ ^

1 ^^s.^ •~~- ^ ^~~~^ ~ -,

^^"^^^~~^-—^-IILT ^ ^
' 1 1 1 1 I 1 1

Fig. 19. Warping ratio vs. equivalent plate thickness

135
FOURTH QUARTER / 1978
 induced distortional stress does not change, indicating a
rigid diaphragm. Therefore, setting Q = 100 and solving
for
Qb cosa K\S
2Ed^
gives
Sb <^3
Ah = 75 (30)
J2 (d + b)
where
area of one diagonal brace, in.
S diaphragm spacing, in.
d depth of box, in.
b width of box, in.
a angle between diagonal brace and horizontal
Fig. 20. Internal diaphragms (Fig. 21)
web thickness, in.
El
Examination of the induced live load ratio effect gives: E

DESIGN SPECIFICATION
As of this date, the AASHTO specifications pertain only
w^here subscripts LL and DL represent live load and dead to straight box girders. There is, however, a set of specifi-
load, respectively. cations available to the engineer, which have been devel-
Combining Eqs. (27) and (28) and permitting a maxi- oped during a comprehensive project^ ^ called CURT.
mum stress ratio of 10%, required diaphragm spacing is: These specifications relate to both curved I and box girder
/ R \ V2 bridges and refer to the current AASHTO specifications
S<L (,^^, ^ , \ <25ft (29) as needed.
\200L - 7500/ ^ ^ A summary of these proposed specifications is given in
The effect of the diaphragm stiffness, Q, on the induced Table 11. Examination of this table shows the similarity
distortion stress^^ is shown in Fig. 21. For Q > 100, the to the present straight box girder design criteria.

E^d

NU -- No. Diaphragms

T" "n—
10 100

Diaphragm Stiffness Q

Fig. 21. Distortional stress vs. diaphragm stiffness

136
ENGINEERING JOURNAL / AMERICAN INSTITUTE OF STEEL CONSTRUCTION
 Table 11. Summary of Proposed Design Specifications (Ref. 11)

Item Curved I Girder Curved Box Girder

Compression
Flange
(Note: See Table 12 for values of PBPW-)
1 b 1 1

Positive moment region: r "

1
b/t < 4 4 0 0 / V ^ ^ ^ '
Negative moment region:
b/t < {6H0/Fy)X
b/t < AAm/VFy
X = \+-(^-0A5) > 1.0
3\Fy 1

Bottom flange:
/ , = (^t'^W
0 = O.O7A;3„4 forn>l
= 0.125y^3 f o r n = l
k — buckling coefficient < 4
n = number of longitudinal stiffeners
ki),ks = buckling coefficient for compression and shear,
respectively, where 4 > ^^ > 2
For flange (including stiffeners) having an allow^able
stress same as for tension flange:
w 3070VI
- < 7=r-X,
t VFy
X i = 1 for rz = 1
7— 1
= 0.93 + / l . 6 - - \ (—\ > 1.0 for n > 1

5.34 + 2.84^/,/z^;/3
ks = < 5.34
2(n + 1)
w 6650VI
If - > above, but < ^=r— Xo or 60: "T T i
t v ^
^^.J^ ^ J '
F, = \0.326Fy + 0.224F, sm ( . / 2 ) / ^ ^ ^ ^ v ^ ^ ^ - ( V O V ^ X 1 ^
L ^ > / V6650VTA:2-3070 VIXi/J
C = V\- 9.0 (f,/Fy)^

w 6650\/I
If- > X2 , but < 60:
t Fy

Fb= \4.4k{t/w)^ C X 10^

or

L \w/ J \4.4ksHt/w) X 106

whichever is smaller

Tension F^=0.55F,[l-9.2(/,/F,)2]V2
U + / 6 < 0.55 Fy
Flange /^ < 0.337^^

Web Plate For do/R <0.02:

(No longitudinal Use A A S H T O Spec. Art. 1.7.70(A). '^ 1 1 -1
V r>
stiffeners)
For do/R> 0.02:
Jl
D
23,000 \
but not less than £)/170
*
1
yp = 1.19 - \0{do/'R) + 2>4{do/R)^
•— 1 - ^ l >-^
w^here do = actual distance betw^een transverse stiffeners, in.
R = radius of girder curvature, in.

(cont'd next page)

137

FOURTH QUARTER / 1978

 Table 11 (cont'd). Summary of Proposed Design Specifications (Ref. 11)

Item Curved I Girder Curved Box Girder Detail

Web Plate
(Single 46,000 \^p/
longitudinal but not less than /)/340 rr
stiffener)
/^Al/2 /do\
^ = ,-2.9(-) +2.2(-)

AASHTO Spec. Art. 1.7.71 applies, except

Web Plate
/, > doty/\0.92
(Transverse
J = [(25Z)y^2) _ 20]X > 5
intermediate T '—T
stiffen^rs) 1.0 + [(4/Z))-0.78]Z-^
1775
D t —J
when 0.78 < d/D < 1.0 and 0 < Z < 10
d <\\,000 t/\/% (req'd spacing of stiffeners)
Z = 0.95dyRt J- , J
Note: If/ > DV/^/1500, no stiffener req'd.

1 1
Longi-
tudinal
Stiffeners is = 6 ' v ^ / 2 2 5 0
r > G^oV^/23,000
D
r'
v^here _J
i
J 1
Is = moment of inertia of stiffener
r = radius of gyration of stiffener
^8t| k:=, -l-tg
Note: In computing Is and r, a centrally located
web strip < 18/ shall be considered as part
of the longitudinal stiffener.

Shear Pc = force on a connector

Connectors
= IP^ + F2 + 2PF sin - j < 0^^

0 = 0.85
Su = ultimate strength of connector [see AASHTO Spec. Art. 1.7.100 (A) (2)]
P = P/N
N = no. of conn, between pts. of max. pos. moment and
adjacent end supports or dead load points of contraflexure,
or between pts. of max. neg. moment and adjacent
dead load points of contraflexure
P = 0.85/^c ^^ or AsFy [whichever is smaller at pts.
of max. pos. moment. At pts. of neg. moment, see
AASHTO Spec. Art. 1.7.100 (A)(2)]
P{\ - cos<9)
~ 4KNs sin(^/2)
6 = angle subtended between pt. of max. moment (pos. or neg.)
and adjacent pt. of contraflexure or support

A: = 0.166 ( 1^ +0.375

As = total area of steel section, incl. cover plates

138

ENGINEERING JOURNAL / AMERICAN INSTITUTE OF STEEL CONSTRUCTION

 Table 12 . Curvature R e d u c t i o n F a c t o r PBPW for Allowable Stress

l/b
/ fw r
R fb \ 7 8 9 10 1 12 14 16 18 20 22 24

0.50 0.74 0.75 0.75 0.75 0.75 0.76 0.76 0.76 0.77 0.77 0.77
0.25 0.84 0.84 j 0.35 0.85 0.85 0.85 0.86 0.86 0.87 0.87 0.87
0.008 0.00 0.95 0.94 0.93 0.93 0.91 0.90 0.89 0.87 0.86 0.85 0.84
-0.25 0.77 0.77 0.76 0.76 0.75 0.75 0.74 0.73 0.73 0.72 0.72
-0.50 0.65 0.65 0.65 0.65 0.64 0.64 0.64 0.63 0.63 0.63 0.63

0.50 0.74 0.74 0.74 0.74 0.74 0.74 0.74 0.74 0.74 0.75 0.75
0.25 0.83 0.83 0.83 0.83 0.84 0.84 0.84 0.84 0.84 0.84 0.84
0.010 0.00 0.93 0.93 0.92 0.91 0.89 0.88 0.86 0.85 0.83 0.82 0.81
-0.25 0.76 0.76 0.75 0.75 0.74 0.73 0.72 0.71 0.70 0.70 0.69
-0.50 0.64 0.64 0.64 0.63 0.63 0.62 0.62 0.61 0.61 0.61 0.60

0.50 0.72 0.72 0.72 0.72 0.71 0.71 0.71 0.71 0.71 0.70 0.70
0.25 0.81 0.81 0.81 0.81 0.81 0.80 0.80 0.80 0.80 0.80 0.79
0.014 0.00 0.91 0.90 0.89 0.88 0.86 0.84 0.82 0.80 0.78 0.76 0.75
-0.25 0.74 0.74 0.73 0.72 0.71 0.69 0.68 0.67 0.66 0.65 0.64
-0.50 0.63 0.62 0.62 0.61 0.60 0.59 0.59 0.58 0.57 0.56 0.56

0.50 0.71 0.70 0.70 0.70 0.69 0.69 0.68 0.68 0.67 0.67 0.66
0.25 0.80 0.79 0.79 0.79 0.78 0.78 0.77 0.77 0.76 0.76 0.75
0.018 0.00 0.89 0.87 0.86 0.85 0.82 0.80 0.78 0.76 0.74 0.72 0.70
-0.25 0.72 0.71 0.71 0.70 0.68 0.66 0.65 0.63 0.62 0.61 0.60
-0.50 0.61 0.60 0.60 0.59 0.58 0.57 0.56 0.55 0.54 0.53 0.52

0.50 0.69 0.69 0.68 0.68 0.67 0.66 0.66 0.65 0.64 0.64 0.63
0.25 0.78 0.78 0.77 0.77 0.76 0.75 0.74 0.73 0.73 0.72 0.71
0.022 0.00 0.87 0.85 0.83 0.82 0.79 0.76 0.74 0.72 0.69 0.67 0.65
-0.25 0.71 0.70 0.68 0.67 0.65 0.64 0.62 0.60 0.59 0.57 0.56
-0.50 0.60 0.59 0.58 0.57 0.56 0.54 0.53 0.52 0.51 0.50 0.49

0.50 0.68 0.67 0.67 0.66 0.65 0.64 0.63 0.63 0.62 0.61 0.60
0.25 0.77 0.76 0.76 0.75 0.74 0.73 0.72 0.71 0.70 0.69 0.68
0.026 0.00 0.85 0.83 0.81 0.79 0.76 0.73 0.71 0.68 0.66 0.64 0.62
-0.25 0.69 0.68 0.66 0.65 0.63 0.61 0.59 0.57 0.56 0.54 0.53
-0.50 0.58 0.57 0.56 0.55 0.54 0.52 0.51 0.49 0.48 0.47 0.46

0.50 0.67 0.66 0.65 0.65 0.64 0.62 0.61 0.60 0.60 0.59 0.58
0.25 0.76 0.75 0.74 0.73 0.72 0.71 0.69 0.68 0.67 0.66 0.66
0.030 0.00 0.83 0.81 0.79 0.77 0.74 0.70 0.68 0.65 0.63 0.60 0.58
-0.25 0.67 0.66 0„65 0.63 0.61 1 0.59 0.56 0.55 0.53 0.51 0.50
-0.50 0.57 0.56 0.55 0.54 0.52 0.50 0.48 0.47 0.46 0.45 0.43

0.50 0.64 0.63 0.62 0.62 0.60 0.59 0.57 0.56 0.55 0.54 0.53
0.25 0.73 0.72 0.71 0.70 0.68 0.66 0.65 0.64 0.62 0.61 0.60
0.040 0.00 0.78 0.76 0.74 0.71 0.68 0.64 0.61 0.58 0.56 0.53 0.51
-0.25 0.64 0.62 0.60 0.59 0.56 0.53 0.51 0.49 0.47 0.45 0.44
-0.50 0.54 0.52 0.51 0.50 0.48 0.46 0.44 0.42 0.41 0.39 0.38

0.50 0.62 0.61 0.60 0.59 0.47 0.56 0.54 0.53 0.52 0.51 0.50
0.25 0.70 0.69 0.68 0.67 0.65 0.63 0.61 0.60 0.59 0.58 0.55
0.050 0.00 0.74 0.71 0.69 0.67 0.63 0.59 0.56 0.53 0.50 0.48 0.45
-0.25 0.60 0.58 0.57 0.55 0.52 0.49 0.46 0.44 0.42 0.40 0.39
-0.50 0.51 0.49 0.48 0.47 0.44 0.42 0.40 0.38 0.37 0.35 0.34

{cont'd next page)

139
FOURTH QUARTER / 1978
 Table 12 (cont'd).
l/b
I fw
R fb 7 8 9 10 12 14 16 18 20 22 24

0.50 0.60 0.59 0.58 0.57 0.55 0.53 0.52 0.51 0.50 0.49 0.48
0.25 0.68 0.67 0.66 0.64 0.62 0.60 0.59 0.57 0.56 0.52 0.49
0.060 0.00 0.70 0.68 0.65 0.63 0.58 0.54 0.51 0.48 0.45 0.43 0.41
-0.25 0.57 0.55 0.53 0.51 0.48 0.45 0.43 0.40 0.38 0.37 0.35
-0.50 0.48 0.47 0.45 0.44 0.41 0.39 0.37 0.35 0.33 0.32 0.31

0.50 0.59 0.57 0.56 0.55 0.53 0.52 0.50 0.49 0.48 0.47 0.46
0.25 0.66 0.65 0.64 0.62 0.60 0.58 0.57 0.55 0.51 0.48 0.45
0.070 0.00 0.67 0.64 0.61 0.59 0.54 0.51 0.47 0.44 0.42 0.39 0.37
-0.25 0.55 0.52 0.50 0.48 0.45 0.42 0.39 0.37 0.35 0.33 0.32
-0.50 0.46 0.44 0.43 0.41 0.38 0.36 0.34 0.32 0.30 0.29 0.28

0.50 0.57 0.56 0.55 0.53 0.51 0.50 0.48 0.47 0.46 0.45 0.44
0.25 0.65 0.63 0.62 0.60 0.58 0.56 0.55 0.51 0.47 0.44 0.41
0.080 0.00 0.64 0.61 0.58 0.56 0.51 0.47 0.44 0.41 0.38 0.36 0.34
-0.25 0.52 0.50 0.48 0.46 0.42 0.39 0.37 0.34 0.33 0.31 0.29
-0.50 0.44 0.42 0.40 0.39 0.36 0.34 0.31 0.30 0.28 0.27 0.26

0.50 0.56 0.54 0.53 0.52 0.50 0.48 0.46 0.45 0.44 0.43 0.42
0.25 0.63 0.61 0.60 0.58 0.56 0.54 0.51 0.47 0.44 0.41 0.38
0.090 0.00 0.61 0.58 0.55 0.53 0.48 0.44 0.41 0.38 0.36 0.34 0.32
-0.25 0.50 0.48 0.45 0.43 0.40 0.37 0.34 0.32 0.30 0.29 0.27
-0.50 0.42 0.40 0.38 0.37 0.34 0.31 0.29 0.28 0.26 0.25 0.24

0.50 0.54 0.52 0.51 0.49 0.47 0.45 0.44 0.43 0.41 0.40 0.40
0.25 0.61 0.59 0.57 0.56 0.53 0.51 0.48 0.44 0.41 0.38 0.35
0.100 0.00 0.59 0.56 0.53 0.50 0.45 0.42 0.38 0.36 0.33 0.31 0.29
-0.25 0.48 0.45 0.43 0.41 0.38 0.35 0.32 0.30 0.28 0.27 0.25
-0.50 0.40 0.38 0.37 0.35 0.32 0.30 0.28 0.26 0.24 0.23 0.22

COMPUTER SOLUTION EI^ , EI^ + GKT „ ^.

-^V'" +^ V" - EI^4>'"
Theory—The general load deformation response of a
curved girder, which may have an arbitrary geometry when FJ
subjected to combined vertical {qy), lateral {qx), longitu- + GKrr-^cp + m, = 0 (32)
dinal iqz), and moment (m^) uniform loads applied along
the girder, as shown in Fig. 22, is given by the following where
differential equations, as developed by Vlasov.^^ rj = vertical deflection along girder (};-axis)
GKT ff EI^u 0 = transverse rotation of girder (about z-axis)
EIx = primary bending stiffness
GKT ~ torsional stiffness
+^ r + qy = 0 (31) Elyj = warping stiffness
R = radius to center line of box

Fig. 22. Box girder forces Fig. 23. Box girder displacements

140
ENGINEERING JOURNAL / AMERICAN INSTITUTE OF STEEL CONSTRUCTION
 hi 2S^GKT FI
2
^^•¥
7?2 ^ ^EI
^-
/?2
•-qyA' (33)
t:iu, /A7,A 2A-'^(A7, + GKi ^
R \ R ) R /?

LL RR

EI^, Eh,\ , 2\\EIu~^GKi)

R Kf)-^»iF^] (f) -['(f)*"»ir^] R
= m^A4 (34)

6(A7„,) + 2A2GA'/ +
- | 4 ( i ' ; 4 , ) + A2GA^7 /?2 -|4(A7,„) +A2GAVI EL,
EK,

Figure 24

The displacements of the box are shown in Fig. 23. of such a box is given by the following differential equation,
These equations have been solved simultaneously by as developed by Dabrowski:^
writing the differentials in central finite difference form."^'-^
The mesh spacing between nodes is defined as A, resulting ylV J^ 4 ^ 4 ^ (39)
in Eqs. (33) and (34), as shown in Fig. 24. The general WA''
mesh patterns are appropriately modified to accommodate where
exterior (simple or fixed) supports in bending or torsion and
interior supports. y = angular distortion of box section
The deformations rj and 0 at each node are then used to M , = internal primary bending moment [Eq.
(35)]
evaluate internal forces, as given by Eqs. (35) through (38),
R radius to center line of box
defined by Vlasov:^^
WA* p, X geometric properties of box
external torsional loading per length (as
M. = -EI, (T," - I) (35)
previously described)
The solution of Eq. (39) can readily be performed by
Bt = -EL, (36) converting the differential into finite difference form, giving
Eq. (40):

(37) (l)(-4)(6 + 4XW)(-4)(1) = 7 ^ (p f" + Y )

WA^
(40)
M„, = -EL
(*-f) (38)
This equation is appropriately modified to consider interior
diaphragms or supports by assuming 7 = 0. In evaluating
/here Eq. (40), the internal bending moment M^ is required in
primary bending moment addition to the external torsional moment m^. Thus, the
Bt bimoment (warping normal stress function) primary analysis is first conducted to determine M^; then
MsT pure torsional moment the distortional analysis is performed. If interior transverse
warping moment diaphragms are imposed, thus inhibiting distortional be-
havior, 7 at that location is assumed to be zero.
Equations (35) through (38) have also been written in finite The induced normal stress due to distortion of the cross
difference notation, thus permitting evaluation of the forces section is given by:
upon determination of deformations.
The primary normal stresses induced into the box are oy= - 7 M * . W (41)
then computed by the classic equations given in the section
where
"Bending and Torsional Stresses."
7^^ = second derivative of angular distortion
Distortion of Curved Sections—In the previous section, the
computed in difference form
load-deformation equations were developed on the premise
A*, W = cross-sectional properties
that the box cross section retains its shape. As described in
the discussion of ''Distortional Stresses", it is known that All of the above equations have been incorporated into
the cross section distorts. The load-deformation response a computer program which will now be described.

141
FOURTH QUARTER / 1978
 Bending moment, kip-in.
Shear force, kips
St. Venant torque, kip-in.
W a r p i n g moment, kip-in.
Total torque, kip-in.
Fig. 25. Nodal point configurations Bimoment, kip-in.^
Angular distortion, rad
Normal bending stresses, ksi
Normal warping stresses, ksi
Computer Program—As has been previously mentioned, 1. Normal distortional stresses, ksi
the governing differential equation for cross-sectional A new version of the box beam program is being devel-
distortion w^as solved using the finite difference numerical oped and will accommodate continuous spans, automatic
technique. Also noted was the need for the values of the dead and live load generation, influence lines, force enve-
independent variable M^ prior to solving Eq. (39); there- lopes ( M , F , Bi, T) and selection of plate size according
fore, the equations for torsion and bending of curved girders to the specifications given in Table 11.
must first be determined. T h i s w^as done using an existing
computer program,^ which utilizes the finite difference REFERENCES
method to solve the differential equations, as previously
described. Once the values for the vertical displacements, 1. Hall, D. H. et al. Horizontally Curved Steel Box Girder
r;, and the rotations, 0, have been obtained, they and their Bridges—A Survey Task Committee on Curved Box
appropriate derivatives are substituted into Eq. (35) to Girders, ASCE-AASHTO Committee on Flexural Members,
Committee on Metals, Structural Division, ASCE {to be
determine the values for the induced bending moments. T h e
published).
values for the bending moments are then transferred to a
2. American Association of State Highway and Transportation
subroutine that has been written to carry out the distor- Officials Standard Specifications for Highway Bridges
tional analysis. Finally, from this subroutine the values for 11th Edition, 1973.
the distortional stresses are obtained. 3. Mattock, A. H. and S. B. Johnston An Experimental and
A general schematic of a curved beam used in the com- Analytical Study of the Lateral Distribution of Load in
puter program^'^ is shown in Fig. 25. Composite Box Girder Bridges Final Report, U.S.S. Corp.,
T h e computer program stores the banded coefficient Pittsburgh, Pa., 1968.
array in a rectangular matrix to economize on the storage 4. Heins, C. P. and J. C. Oleinik Curved Box Beam Bridge
Analysis Computers and Structures Journal, Vol. 6, pp.
locations required. T h e resulting augmented matrix is
65-73, Pergamon Press, London, 1976.
solved using a decomposition method. If a diaphragm is to
5. Oleinik, J. C. and C. P. Heins Diaphragms for Curved Box
be located at an arbitrary node point z, the angular distor- Beam Brid,2;es ASCE Journal of the Structural Division, Vol.
tion at this point is assumed to be equal to zero. In order to 101,No.ST10,Oct. 1975.
make the equations compatible with this assumption, the 6. Oleinik, J. C and C P. Heins Diaphragm Spacing Re-
^th j.Q^ ^j^j corresponding diagonal of the coefficient array, quirements for Curved Steel Box Beam Bridges Civil En-
as well as the i^^ term of the load array, must be eliminated gineering Report No. 58, University of Maryland, College
from the augmented matrix.- After the appropriate terms Park, Md., Aug. 1974.
have been eliminated, the augmented matrix is then con- 1. Heins, C P. Bending and Torsional Design of Structural
densed. T h e above procedure was performed at each dia- Members Lexington Books, Lexington, Mass., 1975.
phragm location before the equations were solved for the 8. Nakai, H. and C P. Heins Analysis Criteria for Curved
Bridges ASCE Journal of the Structural Division, Vol. 103,
unknown angular distortions between the diaphragms.
No. ST7, July 1977.
Program Input- - T h e general computer input data in- 9. Dabrowski, R. Curved Thin Walled Girders Cement and
eluded: Concrete Association, London, England 1968.
10. Williamson, D. M. BEF Analysis for Cross-Sectional
a. N u m b e r of mesh points Deformation of Curved Box Girders with Internal Dia-
b. Support conditions, number of interior supports phragms M.S. THESIS= Civil Engr. Dept., Carnegie-
c. Material properties Mellon University, Pittsburgh, Pa., April 1974.
d. Girder geometry 11. Tentative Design Specifications for Horizontally Curved
e. Radius Highway Bridges Final Report, CURT—Project
HPR2-011), March, 1975.
f. Loading (uniform or concentrated)
12. Mozer, J. D. and C. G. Culver Behavior of Curved Box
Program Output—The program will print out the fol- Girders—Laboratory Observations ASCE Specialty
lowing for each mesh-point along the girder length. Conference on Metal Bridges, St. Louis, Mo., Nov. 12-13,
1974.
a. Vertical deflection, in. 13. Vlasov, V. Thin Walled Beam Theory NSF, Washington,
b. Rotation, rad D.C, 1961.

142
ENGINEERING JOURNAL / AMERICAN INSTITUTE OF STEEL CONSTRUCTION