The National Academies Press: Proposed Modification To AASHTO Cross-Frame Analysis and Design (2021)

THE NATIONAL ACADEMIES PRESS
This PDF is available at http://nap.edu/26074 SHARE

   
Proposed Modification to AASHTO Cross-Frame Analysis and

Design (2021)
DETAILS
156 pages | 8.5 x 11 | PAPERBACK

ISBN 978-0-309-67373-0 | DOI 10.17226/26074
CONTRIBUTORS
GET THIS BOOK Matthew Reichenbach, Joshua White, Sunghyun Park, Esteban Zecchin, Matthew
Moore, Yangqing Liu, Chen Liang, Bal zs K vesdi, Todd Helwig, Michael
Engelhardt, Robert Connor, Michael Grubb, Ferguson Structural Engineering
FIND RELATED TITLES Laboratory, The University of Texas at Austin; National Cooperative Highway
Research Program; Transportation Research Board; National Academies of
SUGGESTED CITATION
Sciences, Engineering, and Medicine
National Academies of Sciences, Engineering, and Medicine 2021. Proposed

Modification to AASHTO Cross-Frame Analysis and Design. Washington, DC: The
National Academies Press. https://doi.org/10.17226/26074.

Visit the National Academies Press at NAP.edu and login or register to get:
– Access to free PDF downloads of thousands of scientiﬁc reports

– 10% off the price of print titles
– Email or social media notiﬁcations of new titles related to your interests
– Special offers and discounts

Distribution, posting, or copying of this PDF is strictly prohibited without written permission of the National Academies Press.
(Request Permission) Unless otherwise indicated, all materials in this PDF are copyrighted by the National Academy of Sciences.
Copyright © National Academy of Sciences. All rights reserved.

Proposed Modification to AASHTO Cross-Frame Analysis and Design
N AT I O N A L C O O P E R AT I V E H I G H W AY R E S E A R C H P R O G R A M
NCHRP RESEARCH REPORT 962

Proposed Modification
to AASHTO Cross-Frame
Analysis and Design
Matthew Reichenbach
Joshua White
Sunghyun Park
Esteban Zecchin
Matthew Moore
Yangqing Liu
Chen Liang
Balázs Kövesdi
Todd Helwig
Michael Engelhardt
Robert Connor
Michael Grubb
Ferguson Structural Engineering Laboratory
The University of Texas at Austin
Austin, Texas
Subscriber Categories
Bridges and Other Structures
Research sponsored by the American Association of State Highway and Transportation Officials
in cooperation with the Federal Highway Administration
2021
Copyright National Academy of Sciences. All rights reserved.

NATIONAL COOPERATIVE HIGHWAY NCHRP RESEARCH REPORT 962

RESEARCH PROGRAM
Systematic, well-designed, and implementable research is the most Project 12-113
effective way to solve many problems facing state departments of ISSN 2572-3766 (Print)
transportation (DOTs) administrators and engineers. Often, highway ISSN 2572-3774 (Online)
problems are of local or regional interest and can best be studied by ISBN 978-0-309-67373-0
state DOTs individually or in cooperation with their state universities Library of Congress Control Number 2020952857
and others. However, the accelerating growth of highway transporta-
© 2021 National Academy of Sciences. All rights reserved.
tion results in increasingly complex problems of wide interest to high-
way authorities. These problems are best studied through a coordinated
program of cooperative research.
Recognizing this need, the leadership of the American Association COPYRIGHT INFORMATION
of State Highway and Transportation Officials (AASHTO) in 1962 ini- Authors herein are responsible for the authenticity of their materials and for obtaining
tiated an objective national highway research program using modern written permissions from publishers or persons who own the copyright to any previously
scientific techniques—the National Cooperative Highway Research published or copyrighted material used herein.
Program (NCHRP). NCHRP is supported on a continuing basis by Cooperative Research Programs (CRP) grants permission to reproduce material in this
funds from participating member states of AASHTO and receives the publication for classroom and not-for-profit purposes. Permission is given with the
full cooperation and support of the Federal Highway Administration understanding that none of the material will be used to imply TRB, AASHTO, FAA, FHWA,
FTA, GHSA, NHTSA, or TDC endorsement of a particular product, method, or practice.
(FHWA), United States Department of Transportation, under Agree-
It is expected that those reproducing the material in this document for educational and
ment No. 693JJ31950003. not-for-profit uses will give appropriate acknowledgment of the source of any reprinted or
The Transportation Research Board (TRB) of the National Academies reproduced material. For other uses of the material, request permission from CRP.
of Sciences, Engineering, and Medicine was requested by AASHTO to
administer the research program because of TRB’s recognized objectivity
and understanding of modern research practices. TRB is uniquely suited
NOTICE
for this purpose for many reasons: TRB maintains an extensive com-
mittee structure from which authorities on any highway transportation The research report was reviewed by the technical panel and accepted for publication
according to procedures established and overseen by the Transportation Research Board
subject may be drawn; TRB possesses avenues of communications and
and approved by the National Academies of Sciences, Engineering, and Medicine.
cooperation with federal, state, and local governmental agencies, univer-
sities, and industry; TRB’s relationship to the National Academies is an The opinions and conclusions expressed or implied in this report are those of the
researchers who performed the research and are not necessarily those of the Transportation
insurance of objectivity; and TRB maintains a full-time staff of special- Research Board; the National Academies of Sciences, Engineering, and Medicine; the
ists in highway transportation matters to bring the findings of research FHWA; or the program sponsors.
directly to those in a position to use them.
The Transportation Research Board; the National Academies of Sciences, Engineering,
The program is developed on the basis of research needs iden- and Medicine; and the sponsors of the National Cooperative Highway Research Program
tified by chief administrators and other staff of the highway and do not endorse products or manufacturers. Trade or manufacturers’ names appear herein
transportation departments, by committees of AASHTO, and by solely because they are considered essential to the object of the report.
the FHWA. Topics of the highest merit are selected by the AASHTO
Special Committee on Research and Innovation (R&I), and each year
R&I’s recommendations are proposed to the AASHTO Board of Direc-
tors and the National Academies. Research projects to address these
topics are defined by NCHRP, and qualified research agencies are
selected from submitted proposals. Administration and surveillance of
research contracts are the responsibilities of the National Academies
and TRB.
The needs for highway research are many, and NCHRP can make
significant contributions to solving highway transportation problems
of mutual concern to many responsible groups. The program, however,
is intended to complement, rather than to substitute for or duplicate,
other highway research programs.
Published research reports of the
NATIONAL COOPERATIVE HIGHWAY RESEARCH PROGRAM

are available from
Transportation Research Board

Business Office
500 Fifth Street, NW
Washington, DC 20001
and can be ordered through the Internet by going to

https://www.nationalacademies.org
and then searching for TRB
Printed in the United States of America

The National Academy of Sciences was established in 1863 by an Act of Congress, signed by President Lincoln, as a private, non-
governmental institution to advise the nation on issues related to science and technology. Members are elected by their peers for
outstanding contributions to research. Dr. Marcia McNutt is president.
The National Academy of Engineering was established in 1964 under the charter of the National Academy of Sciences to bring the
practices of engineering to advising the nation. Members are elected by their peers for extraordinary contributions to engineering.
Dr. John L. Anderson is president.
The National Academy of Medicine (formerly the Institute of Medicine) was established in 1970 under the charter of the National
Academy of Sciences to advise the nation on medical and health issues. Members are elected by their peers for distinguished contributions
to medicine and health. Dr. Victor J. Dzau is president.
The three Academies work together as the National Academies of Sciences, Engineering, and Medicine to provide independent,
objective analysis and advice to the nation and conduct other activities to solve complex problems and inform public policy decisions.
The National Academies also encourage education and research, recognize outstanding contributions to knowledge, and increase
public understanding in matters of science, engineering, and medicine.
Learn more about the National Academies of Sciences, Engineering, and Medicine at www.nationalacademies.org.
The Transportation Research Board is one of seven major programs of the National Academies of Sciences, Engineering, and Medicine.
The mission of the Transportation Research Board is to provide leadership in transportation improvements and innovation through
trusted, timely, impartial, and evidence-based information exchange, research, and advice regarding all modes of transportation. The
Board’s varied activities annually engage about 8,000 engineers, scientists, and other transportation researchers and practitioners from
the public and private sectors and academia, all of whom contribute their expertise in the public interest. The program is supported by
state transportation departments, federal agencies including the component administrations of the U.S. Department of Transportation,
and other organizations and individuals interested in the development of transportation.
Learn more about the Transportation Research Board at www.TRB.org.

COOPERATIVE RESEARCH PROGRAMS
CRP STAFF FOR NCHRP RESEARCH REPORT 962

Christopher J. Hedges, Director, Cooperative Research Programs
Lori L. Sundstrom, Deputy Director, Cooperative Research Programs
Waseem Dekelbab, Senior Program Officer
Tyler Smith, Senior Program Assistant
Eileen P. Delaney, Director of Publications
Natalie Barnes, Associate Director of Publications
Janet M. McNaughton, Senior Editor
NCHRP PROJECT 12-113 PANEL

Field of Design—Bridges
Norman L. McDonald, Ames, IA (Chair)
Xiaohua Hannah Cheng, New Jersey Department of Transportation, Trenton, NJ
Domenic A. Coletti, HDR, Raleigh, NC
Karl H. Frank, Consultant, Austin, TX
John S. Hastings, National Steel Bridge Alliance (NSBA), Nashville, TN
Thomas P. Macioce, Pennsylvania Department of Transportation, Harrisburg, PA
Michelle Lauren Romage-Chambers, Texas Department of Transportation, Austin, TX
Dayi Wang, FHWA Liaison
Stephen F. Maher, TRB Liaison
AUTHOR ACKNOWLEDGMENTS
The research reported herein was performed under NCHRP Project 12-113 by the Department of
Civil, Architectural, and Environmental Engineering at the University of Texas at Austin (UT Austin).
UT Austin was the contractor for this study.
The Texas Department of Transportation (TxDOT) provided substantial assistance to the research
team throughout the field experimental program. TxDOT organized lane closures and provided dump
trucks for controlled live load tests as well as general traffic control to assist in the instrumentation of
the bridges. The research team specifically acknowledges Dr. Ken Lin, Lynn Champagne, Chris Baker,
David Jeffreys, Adam Galland, Melody Galland, Maria Aponte, Ray Castillo, Jaime Castaneda, and
Larry Whittington for their contributions to the field studies. Significant assistance was also provided
by Jamie Farris, Dennis Johnson, and Walter Fisher in identifying bridges for possible instrumentation.
Texas Advanced Computing Center (TACC) provided computing resources to conduct the extensive
parametric studies associated with this research. The research team specifically acknowledges Bob Garza
and Dr. Ritu Arora for their assistance throughout the process.
The research team thanks Dr. Tom Murphy, Dr. John Kulicki, and Dr. Andrzej Nowak for informa-
tion provided from the SHRP 2 R19B study that was invaluable for understanding weigh-in-motion data
gathered and related reliability studies.

FOREWORD
By Waseem Dekelbab
Staff Officer
NCHRP Research Report 962 presents proposed AASHTO bridge design specifications
for the analysis and design of cross-frames in straight and curved steel I-girder bridges.
The proposed specifications are based on comprehensive analytical and testing programs
for investigating the effects of support skew on the cross-frame behavior, fatigue design
forces, the strength and stiffness requirements for stability bracing, and the influence of
cross-frame member end connection upon the cross-frame stiffness. This report will be of
immediate interest to bridge engineers.
Developments in bridge design and analysis in recent years have created the need for
improvements to cross-frame analysis and design for steel girder bridges. In the past, the
configuration of cross-frame systems was generally based upon standard designs in which
member sizes and layouts were dependent upon geometry and minimum member cross-
section requirements. The opportunities for improvements to cross-frame analysis and
design cover a variety of topics including: (1) improved definition of fatigue loading for
cross-frames in curved and/or severely skewed steel girder bridges, analyzed using refined
analysis methods; (2) implementation of stability bracing strength and stiffness require-
ments in the context of AASHTO Load and Resistance Factor Design (LRFD) bridge
design; and (3) additional guidance for adjustment of the effective stiffness of cross-frame
members in refined analysis models to reflect the influence of end connections on cross-
frame member stiffness. Addressing these topics could result in a dramatic improvement in
reliability and economy of cross-frames for steel I-girder bridges.
Research was performed under NCHRP Project 12-113, “Proposed Modification to
AASHTO Cross-Frame Analysis and Design” by the University of Texas at Austin and
included: (1) controlled live load tests and measurement of in-service stress cycle counts
for three steel I-girder bridges and (2) an extensive analytical parametric study. The results
of the live load tests and analytical study were used to develop the proposed modifica-
tion to the AASHTO specifications.
Several deliverables, provided as appendices, are not included in the report itself but
are available on the TRB website at trb.org by searching for NCHRP Research Report 962.
The appendices are as follows:
• Appendix B: Cross-Frame Design Example (Straight Bridge)
• Appendix C: Cross-Frame Design Example (Curved Bridge)
• Appendix D: Phase I Summary
• Appendix E: Phase II Summary
• Appendix F: Phase III Summary


CONTENTS
1 Summary
4 Chapter 1 Background
4 1.1 Problem Statement and Current Knowledge
5 1.1.1 Best Practices for Cross-Frame Layout and Detailing
6 1.1.2 Cross-Frame Fatigue Behavior
7 1.1.3 Cross-Frame Analysis
7 1.1.4 Cross-Frame Stability Bracing Requirements
8 1.2 Research Objectives
10 1.3 Scope of Study
10 1.4 Organization of Report
12 Chapter 2 Research Approach

15 2.1 Industry Survey
16 2.2 Field Experimental Program and Model Validation
17 2.2.1 Instrumented Bridge Overview
17 2.2.2 Experimental Testing Program Overview
19 2.3 Fatigue Loading Study
20 2.3.1 General Methodology
28 2.3.2 AASHTO Design Loads
31 2.3.3 WIM Records
38 2.4 R-Factor Study (3D Analysis)
39 2.4.1 General Load-Induced Cross-Frame Behavior
40 2.4.2 Stiffness Modification Approach
41 2.4.3 Proposed Eccentric-Beam Approach
43 2.4.4 Parametric Study Overview
46 2.5 Commercial Design Software Study (2D Analysis)
47 2.5.1 General Modeling Assumptions
48 2.5.2 Equivalent Beam Approach for Cross-Frames
49 2.5.3 Approaches for Improving 2D Analyses
52 2.5.4 Parametric Study Overview
52 2.6 Stability Study
56 2.6.1 Buckling in the Composite Condition
57 2.6.2 Bracing Strength Study
59 2.6.3 Bracing Stiffness Study

60 Chapter 3 Findings and Applications

60 3.1 Field Experimental Program and Model Validation
61 3.1.1 Controlled Live Load Test and Model Validation
67 3.1.2 In-Service Monitoring
70 3.2 Fatigue Loading Study
71 3.2.1 Influence of Bridge Geometry
83 3.2.2 Cross-Frame-Specific Load Factors
89 3.2.3 Multiple Presence Study
92 3.2.4 Major Outcomes
94 3.3 R-Factor Study (3D Analysis)
94 3.3.1 Panel-Level Studies (Noncomposite)
96 3.3.2 System-Level Studies (Composite)
103 3.4 Commercial Design Software Study (2D Analysis)
104 3.4.1 Sample Influence-Line Results
106 3.4.2 Postprocessing Error
115 3.4.3 Parametric Study
118 3.4.4 Alternative 2D Postprocessing Procedure
120 3.5 Stability Study
120 3.5.1 Buckling in the Composite Condition
123 3.5.2 Bracing Strength Study
128 3.5.3 Bracing Stiffness Study
131 3.5.4 Consideration of Stability Forces in Design
134 Chapter 4 Conclusions and Suggested Research

134 4.1 Major Conclusions
138 4.2 Suggestions for Implementation
142 4.3 Future Research Needs
143 References
A-1 Appendix A Proposed Modifications to AASHTO LRFD
Note: Photographs, figures, and tables in this report may have been converted from color to grayscale for printing.
The electronic version of the report (posted on the web at www.trb.org) retains the color versions.

SUMMARY
Proposed Modification
to AASHTO Cross-Frame
Analysis and Design
Cross-frames are important structural components that serve many functions throughout
the service life of steel I-girder bridges. They primarily act as stability braces to enhance
the lateral-torsional buckling resistance of the bridge girders during erection and deck
construction. Among other functions, they also distribute live loads to girders across
the bridge width in the final composite condition. Under repetitive load cycles caused
by heavy truck passages, cross-frames and their connections can be susceptible to load-
induced fatigue cracking if not properly designed. Despite recent changes to design speci-
fications that signal cross-frames as potential fatigue-sensitive details given substantial
live load force effects, little load-induced fatigue cracking has been observed in existing
steel I-girder bridges across the country.
Cross-frames have historically been detailed and fabricated based on general rules-of-
thumb and experience. In recent years, however, developments in bridge design specifica-
tions have necessitated updates to cross-frame design and analysis practices. Cross-frames
are now designed and detailed based on rational analysis for all stages of construction and
service life, which has further emphasized the importance of accurate and reliable analysis
techniques and design criteria.
Although considerable research over the past several decades has improved cross-
frame design and analysis, bridge designers have generally lacked adequate guidance
on the appropriate (i) fatigue loading criteria, (ii) analysis procedures, and (iii) stability
bracing requirements for cross-frame systems. With that in mind, NCHRP Project 12-113,
“Proposed Modification to AASHTO Cross-Frame Analysis and Design,” was under-
taken to address many of these knowledge gaps in an attempt to improve the reliability
and economy of cross-frames in steel I-girder bridges. The fundamental objectives of
NCHRP Project 12-113 were to produce quantitatively based methodologies and design
guidelines for the following items:
a. Improved definition of fatigue loading for cross-frames in skewed steel I-girder bridges
(straight or curved) analyzed using refined analysis methods;
b. Investigation of the influence of girder spacing, cross-frame stiffness and spacing
(including staggered layouts), and deck thickness on the force effects in cross-frame
systems;
c. Additional guidance on how to evaluate the influence of end connections on cross-frame
member stiffness in refined analysis models;
d. Evaluation of commercial software programs and their ability to accurately predict
cross-frame forces for various bridge geometries and cross-frame configurations; and
e. Evaluation of stability bracing strength and stiffness requirements for implementation in
the AASHTO Load and Resistance Factor Design (LRFD) Bridge Design Specifications.
1

2 Proposed Modification to AASHTO Cross-Frame Analysis and Design
The objectives were systematically addressed through a series of field experiments, para-
metric finite element analyses, and weigh-in-motion studies (considering high-resolution
data from 16 sites across the country). This report summarizes the results of the research
effort undertaken as part of this project.
An important step in the research was to first validate the computational models
utilizing field data from representative bridges. As such, three steel I-girder bridges of
varying geometry (i.e., a straight bridge with normal supports, a straight bridge with
skewed supports, and a horizontally curved bridge with radial supports) were instru-
mented and experimentally tested in two different ways. First, a controlled live load test
was conducted. Dump trucks of known axle configuration and weight were statically
positioned at different locations on the bridge deck, and the stress response in critical
cross-frame members and girder flanges was subsequently measured. Second, stress cycle
counts induced in the same critical cross-frame members from truck traffic were moni-
tored for a one-month span. The data collected from the controlled live load test were
primarily used to validate a finite element modeling approach. The in-service stress range
data served as a physical benchmark for computational analysis and weigh-in-motion
studies conducted in later phases of the project.
Upon validating a finite element approach, a series of extensive analytical parametric
studies were conducted to expand the breadth and depth of knowledge gained from the
field studies. In general, the load-induced fatigue behavior and stability bracing charac-
teristics of conventional X- and K-type cross-frames were examined for a variety of bridge
geometries commonly found in the United States. These analyses were performed with
different levels of computational refinement, ranging from sophisticated, three-dimensional
(3D) approaches to simplified, two-dimensional (2D) approaches. Based on the results
of the experimental and analytical studies, there were several key findings related to the
five major objectives identified above, as follows:
• The current fatigue loading model implemented in AASHTO LRFD, which was calibrated
based on girder response, does not accurately represent the cross-frame fatigue response
to the U.S. truck spectrum. The current design loads and factors generally overestimate
the force effects in critical cross-frame members when compared to measured field
data or weigh-in-motion records, largely due to the sensitivity of cross-frames to load
position on the bridge deck.
• Load-induced cross-frame forces are highly variable and depend on a number of
parameters. However, the results of this study indicate that significant live load-induced
cross-frame forces are generally correlated with large support skews and tight horizontal
curves. Implementing discontinuous, staggered cross-frame layouts represents a practical
means to mitigate excessive load-induced forces for bridges with large support skews.
In contrast, cross-frame fatigue force effects in straight bridges with little to no support
skews are generally small.
• The load-induced deformational response of cross-frames in composite systems (appli-
cable to in-service bridges) is much different than the deformational response in non-
composite systems (applicable to bridges during construction). Additionally, traditional
3D modeling approaches of cross-frames commonly used by commercial software
programs (i.e., cross-frames treated as pin-ended planar trusses) can misrepresent the
true stiffness of a cross-frame panel and its connections. As such, in-service stiffness
modification factors are proposed to improve analysis procedures for cross-frames.
• Two-dimensional models, which represent cross-frame panels as equivalent beam
elements, are prone to predict erroneous top strut forces and substantial errors in
diagonal member forces when evaluating the composite condition. This is largely

Summary 3
attributed to the assumptions inherent with the postprocessing procedures (i.e., con-
verting equivalent beam forces into cross-frame member forces). Although these dis
crepancies were observed for bridges of all geometries, 2D analysis methods tend to
perform worse for heavily skewed and/or curved bridge systems.
• Previously established bracing stiffness and strength requirements published in the
American Institute of Steel Construction (AISC) Specifications were reviewed and
modified for implementation into the AASHTO LRFD Specifications. Additionally,
stability-related force effects in cross-frames can be combined with other construction-
related load cases via linear superposition. Design examples were developed to demonstrate
these procedures.
Based on the major conclusions of NCHRP Project 12-113, changes to the specification
and commentary language in select AASHTO LRFD Articles are proposed. Design examples
were also developed to support and demonstrate the major findings of this research.
Ballot items were subsequently prepared for consideration by the AASHTO Highway
Subcommittee on Bridges and Structures. In general, the guidance developed from this
study is expected to result in improved reliability and economy of cross-frame design in
steel I-girder bridges.

CHAPTER 1
Background
1.1 Problem Statement and Current Knowledge

Cross-frames serve many roles throughout the construction and service life of steel I-girder
bridges. A major function of cross-frames is to serve as stability braces to enhance the lateral-
torsional buckling (LTB) resistance of the bridge girders. Provided that they are properly
designed and detailed, cross-frames restrain the twist of a girder cross-section at discrete
locations along the length; hence, they are aptly referred to as torsional braces. From a stability
perspective, the critical stage for bracing often occurs during construction of the concrete bridge
deck when the steel girders resist the entire construction load. Once the concrete cures and
the system is composite, stability is not generally critical since the deck provides continuous
lateral and torsional restraint to the top flange. In multi-span bridges, LTB is often perceived
to be critical around interior supports since the bottom flange is in compression; however,
this perceived problem neglects the substantial lateral and torsional restraint provided by the
concrete deck.
Aside from their primary role as stability braces, cross-frames also resist a variety of lateral
and gravity loads throughout the life of a bridge. Cross-frames resist the applied torque on
fascia girders due to typical deck overhang construction and distribute lateral loads across
the structure (e.g., wind). They also restrain differential movement in girders (i.e., vertical
deflection and rotation) caused by dead and construction loads on the noncomposite system
(e.g., externally applied loads and locked-in fit-up forces) and live loads on the composite
system. In horizontally curved bridges, cross-frames are primary members and engage the
girders across the bridge width to behave as a unified structural system and to resist the torsion
developed from the curved geometry.
In the completed structure, the passage of heavy truck traffic often imposes cyclic loading
conditions to cross-frames. Under repeated loading, cross-frames are susceptible to load-
induced fatigue cracking. However, bridge owners have not typically observed extensive
load-induced fatigue cracking over the past few decades.
Historically, cross-frame design has largely been based on experience and general rules-of-
thumb. In 1949, a 25-foot maximum spacing limit was introduced in the AASHO Standard
Specifications for Highway Bridges. Similar to modern practices, most states utilized standard
details and member sizes such that cross-frames were seldom engineered or designed elements.
While the 25-foot requirement generally resulted in satisfactory performance of steel bridge
superstructures, it was essentially an arbitrary limit that sometimes produced overly conser-
vative and uneconomical cross-frame layouts. By the 1980s, spacing limits that resulted in a
large number of cross-frames were recognized as undesirable since (i) cross-frames represent an
expensive component of bridge fabrication and erection, and (ii) regions around cross-frame
panels were found to be susceptible to distortion-induced fatigue cracking.

Background 5
As structural analysis tools and bridge design practices advanced in the following years,
updates to the cross-frame analysis and design procedures became increasingly necessary.
This process was initiated in the 1st Edition of the American Association of State Highway and
Transportation Officials LRFD Bridge Design Specifications (1994). This document, along
with its subsequent updates, is referred to as AASHTO LRFD hereafter. In the 1994 AASHTO
LRFD, the 25-foot spacing limit was removed in favor of a rational design approach. Thus,
rather than using a prescriptive or standard design, the need for cross-frames was to be evaluated
for all stages of construction and in the final condition of the bridge.
In the past two to three decades since the rational design approach was implemented,
considerable research has been conducted to improve the body of knowledge related to cross-
frame behavior in steel bridge systems. Many of these past research findings are documented
in the current 9th Edition of the AASHTO LRFD Specifications (2020). Note that any specific
reference to AASHTO LRFD articles in this report implies the 9th Edition specifications,
unless noted otherwise. Despite those advancements, several gaps in knowledge have been
identified, particularly in the areas of fatigue loading and analysis of cross-frames. Before
identifying the major objectives of NCHRP Project 12-113 in Section 1.2, these past research
advancements and the corresponding updates to AASHTO LRFD are first reviewed.
1.1.1 Best Practices for Cross-Frame Layout and Detailing

With rational analysis in the forefront of modern cross-frame analysis and design, the general
framework for satisfying that requirement is provided in AASHTO LRFD Article 6.7.4.1. The
major functions of cross-frames are outlined, and a distinction in the minimum design require-
ments is made for straight and horizontally curved bridges. For straight bridges with normal
supports, cross-frames have traditionally been designed to only transfer wind loads and meet all
applicable slenderness limits. As such, refined analyses are rarely utilized by designers to obtain
cross-frame force effects in these relatively simple systems.
In skewed bridges, however, load-induced force effects in intermediate cross-frames are more
substantial. The same is true for curved bridges, in which cross-frames are considered primary
members. Therefore, refined analysis models, in which cross-frames are explicitly represented
either by two-dimensional (2D) or three-dimensional (3D) methods, are commonly used for
these more complex superstructures. As specified in Article 6.7.4.1, cross-frames must then be
designed for all applicable limit states at all stages of construction and service, if included in the
structural analysis model used to determine force effects. Thus, the design requirements for
cross-frames in skewed and/or curved systems are more generally comprehensive and rigorous
than those in non-skewed and straight systems.
In terms of general cross-frame layout and details in skewed and curved systems, White et al.
(2015) and Chavel et al. (2016) examined best detailing practices for the fit-up of cross-frames.
Particularly in skewed and curved framing systems, cross-frames can be difficult to fit-up during
erection due to out-of-plumbness and differential girder displacements. Through extensive
analytical studies, detailing guidelines were proposed to ensure reliable fit-up in complex
systems, which have since been incorporated in AASHTO Articles 6.7.2 and 6.7.4.2 and their
corresponding commentary. For instance, the benefits of staggered cross-frame layouts in
heavily skewed bridges to relieve load-induced force effects are outlined.
Aside from this research, additional work has been conducted investigating the stiffness and
fatigue performance of cross-frames in skewed bridges. Quadrato et al. (2014) proposed and
experimentally tested a split-pipe connection detail to replace the bent plate detail commonly
used in skewed cross-frames. Battistini et al. (2013) also investigated alternative cross-frame
designs including the use of tube sections with slotted connection plates to mitigate eccentric

connection effects. While these concepts show promise, neither has been widely adopted in
the steel bridge design and construction industry.
1.1.2 Cross-Frame Fatigue Behavior

Design of cross-frames for fatigue involves consideration of both fatigue loading as well as
fatigue resistance. Recent studies addressing each component are discussed in this subsection.
McDonald and Frank (2009) experimentally tested a variety of cross-frame members and
welded gusset plate connections under cyclic loads. Significant bending deformations were
observed, even under tension, due to the eccentric load path generated by the end connections.
These eccentric loading effects ultimately compromised the load-induced fatigue performance
of the test specimens. In fact, many of the test results indicated that eccentrically loaded single-
angle details ranged from Category E to E′ performance.
Battistini et al. (2013) expanded on McDonald and Frank (2009) by evaluating the fatigue
performance of full cross-frame panels. A variety of typical and proposed connection details
in X- and K-type cross-frame panels were fabricated and tested to failure under cyclic loading
that simulated the relative girder deformations resulting from live load traffic on composite
girder systems. Similar to the previous study, many common details that are prevalent in practice
were shown to have poor load-induced fatigue characteristics due to the presence of localized
stress concentrations near welds and/or unintended bending stresses. From these experimental
studies, typical welded cross-frame connections were designated as a Category E’ detail, the
lowest established level. This work has since been summarized and included in AASHTO LRFD
in Table 6.6.1.2.3-1 (Condition 7.2), beginning with the 2016 Interim Specifications.
Additionally, a substantial amount of research has been conducted on the distortion-
induced fatigue behavior for steel bridge applications spanning several decades (Fisher et al.
1990; Keating et al. 1990; Connor and Fisher 2006; Hartman et al. 2010; Hassel et al. 2013).
Distortion-induced fatigue effects, which were prevalent in cross-frame connections from
the 1950s through the 1980s, are a product of out-of-plane secondary stresses due to small
relative deformations of connected components, typically in localized regions. These stresses
are generally not quantified in analysis nor design. Instead, designers typically address and
mitigate these effects with proper detailing based on the research listed above and the guidance
provided in AASHTO LRFD Article 6.6.1.3. The focus of the present study, however, is on
load-induced fatigue effects.
Although a substantial amount of work has been conducted in recent years to examine the
load-induced fatigue resistance of cross-frames and their connections, the fatigue loading
model has seen much less research. The Strategic Highway Research Program 2 (SHRP 2)
Project R19B study (Modjeski and Masters 2015) recently assessed service limit state design
in bridges. This study resulted in a recalibration of the fatigue load factors in the 8th Edition
AASHTO LRFD (2017) Table 3.4.1.1 to better represent the force effects in girders (i.e., shears
and moments) generated by modern truck traffic as indicated by weigh-in-motion (WIM)
records across the country. The 2015 AASHTO LRFD Interim Specifications also recom-
mended in Article C6.6.1.2.1 that a single truck in a single lane is more appropriate for fatigue
loading of cross-frames than multiple presence effects. The basis for this recommendation,
however, was initially an engineering judgement and visual observation of truck traffic with
unknown weights.
Although comprehensive in terms of girder response, the SHRP 2 R19B study (Modjeski and
Masters 2015) did not explicitly evaluate fatigue loading criteria in the context of cross-frames.
As such, a major objective of NCHRP Project 12-113 was to examine the appropriateness of

Background 7
the current fatigue loading model (i.e., AASHTO LRFD Articles 3.4.1, 3.6.1.4, and 6.6.1.2.1) for
cross-frame analysis and design, as further outlined in Section 1.2.
1.1.3 Cross-Frame Analysis

Rational analysis of cross-frames or any structural element depends on accurately analyzing
the structure and determining the appropriate force effects. Improvements in the analytical
tools over the last few decades allow engineers to carry out sophisticated analyses on bridge
systems that can produce efficient designs satisfying both construction and in-service design
requirements. However, improved analytical refinement generally comes at the cost of increased
computational efforts, which is a general theme of the report herein. As with any analysis,
the accuracy is limited by the modeling assumptions and level of understanding of the funda
mental behavior.
Although 3D models provide the most accurate solutions, they generally require significant
training and experience to use. Two-dimensional and one-dimensional (1D) models, which are
much easier and time-efficient to develop, can at times offer reliable data for simple structures,
particularly for girder forces. AASHTO/NSBA (National Steel Bridge Alliance) G13.1 Guidelines
for Steel Girder Bridge Analysis (2019) comprehensively summarize the limits of simplified
analysis techniques for complex systems with skews and/or horizontal curvature.
While many of these advancements have focused on girder analysis and design, studies
regarding cross-frames have been much more recent. Among other topics, White et al. (2012)
investigated a variety of simplified analysis techniques with respect to cross-frame behavior in
noncomposite bridge systems. In general, it was observed that cross-frame force effects are more
difficult to predict in simplified 2D analyses and that careful consideration by the designer is
required when developing the model.
Even in the most detailed 3D design software packages, cross-frame members are generally
simplified as pin-ended truss elements that do not reflect the impact of eccentric connections.
As such, Wang (2013) and Battistini et al. (2016) explored the use of a simple modification
factor (i.e., a stiffness reduction factor referred to as an R-factor hereafter) in 3D analysis to
more accurately represent the true stiffness of the cross-frame. AASHTO LRFD Articles 4.6.3.1
and 4.6.3.3 document the key outcomes of these past investigations, including the appropriate
equivalent beam approach used in 2D analysis and the modification factor used in 3D models
to account for the effective connection stiffness.
Despite these recent studies, little guidance has been provided on how different analysis
techniques affect the predicted response of cross-frames in composite systems (i.e., cross-
frame response to live loads). Thus, another major objective of NCHRP Project 12-113 was to
examine the limitations of common 3D and 2D analysis methods in the context of cross-frame
forces and AASHTO LRFD Articles 4.6.3.1 and 4.6.3.3. This is further outlined in Section 1.2.
1.1.4 Cross-Frame Stability Bracing Requirements

As noted previously, cross-frames primarily serve as torsional braces to develop the load-
carrying capacity of the bridge girders, especially during steel erection and deck construction.
Like any stability brace, cross-frames must satisfy both stiffness and strength requirements to
adequately brace a girder. Despite this, AASHTO LRFD (2020) currently provides no explicit
guidance on stability bracing requirements. It does, however, address several stability-related
issues in steel I-girder systems, which are tangentially related to cross-frames.
For instance, research conducted by Yura et al. (2008) and Han and Helwig (2017) investi-
gated the system buckling phenomenon in narrow I-girder systems following the collapse of

the Marcy Pedestrian Bridge. Although the Marcy Pedestrian Bridge was a steel tub girder, the
LTB failure is similar to the system buckling mode for which many narrow I-girder systems
are susceptible. AASHTO Article 6.10.3.4.2 summarizes the major findings of these studies
and provides design guidance to estimate the resistance of this system-level failure mode.
Substantial work has also been performed in the area of lean-on bracing (Helwig and Wang
2003, Romage 2008). Lean-on bracing concepts, which have been utilized in building frame
design for decades, were adapted for steel straight I-girder bridge applications. These braces
include only a top and bottom strut and “lean on” adjacent cross-frames to stabilize the
longitudinal girders. Several newly constructed straight skewed I-girder bridges in Texas
have successfully implemented lean-on braces, which subsequently alleviate load-induced force
effects in cross-frames as well as improve fabrication economy by eliminating several full
cross-frame panels.
Despite these recent advancements related to the topic of steel bridge stability, stability
bracing requirements for cross-frames are absent from the AASHTO LRFD Specifications as
of this writing. In contrast, the American Institute of Steel Construction (AISC) Specification
commonly used for building design was the first major specification in the world to include
guidance on stability bracing design. In the current AISC Specification (2016), the stability
bracing provisions are located in Appendix 6, and Article 6.3.2a provides specific strength
and stiffness requirements for torsional point braces such as cross-frames. These require-
ments are largely based on the work conducted by Yura (2001), Helwig and Yura (2015), and
Prado and White (2015). Note that the fundamental work that led to the stability bracing
requirements in the AISC Specification was actually developed from studies on bridge girders
(Yura et al. 1992).
Considering that many of these stability bracing design provisions are applicable for both
building and bridge applications, another major objective of NCHRP Project 12-113 was
to assess the stability bracing guidance in the AISC Specification for implementation into
AASHTO LRFD.
1.2 Research Objectives

As outlined in the preceding section, there has been a concerted effort over the last two
to three decades to improve the rational analysis and design guidelines for cross-frames in
AASHTO LRFD. While these studies have advanced the understanding of the behavior, there
are still significant gaps in knowledge on the analysis, design, and detailing of cross-frames.
These knowledge gaps primarily fall into one of three categories: (i) fatigue loading criteria,
(ii) analysis techniques, and (iii) stability bracing requirements.
With that in mind, the National Cooperative Highway Research Program (NCHRP)
Project 12-113, “Proposed Modification to AASHTO Cross-Frame Analysis and Design,” was
conceived to address many of these knowledge gaps in an attempt to improve the reliability
and economy of cross-frames in steel I-girder bridges. The fundamental objectives of NCHRP
Project 12-113, as identified in the project Request for Proposal (RFP), were to produce quan
titatively based methodologies and design guidelines for the following items:
a. Improved definition of fatigue loading for cross-frames in skewed steel I-girder bridges
(straight or curved) analyzed using refined analysis methods;
b. Investigation of the influence of girder spacing, cross-frame stiffness and spacing (including
staggered layouts), and deck thickness on the force effects in cross-frame systems;

Background 9
c. Additional guidance on how to evaluate the influence of end connections on cross-frame

member stiffness in refined analysis models;
d. Evaluation of commercial software programs and their ability to accurately predict cross-
frame forces for various bridge geometries and cross-frame configurations; and
e. Development of stability bracing strength and stiffness requirements in the context for
implementation in AASHTO LRFD.
Objectives (a) and (b) primarily investigate the fatigue loading characteristics of cross-frames
in composite, finished bridges. Objectives (c) and (d) examine the analysis procedures com-
monly used by designers and commercial bridge software programs, and Objective (e) examines
the stability bracing functions of cross-frames, which are currently absent from AASHTO LRFD.
These five objectives also defined the five major tasks undertaken in this study. The tasks were
addressed herein through a series of field experiments and analytical studies. This report
summarizes the key findings of these studies for practicing engineers. Based on the findings of
these five tasks, the research team developed draft specification and commentary language for
proposed changes to AASHTO LRFD, which are summarized in Chapter 4 and provided in
full in Appendix A. To assist the reader, references to Objectives (a) through (e) are periodi-
cally made in the report hereafter.
For reference, a list of the specific AASHTO LRFD articles recommended for modification
in Appendix A and referenced throughout this report is provided below. For each article or set
of articles, specific questions related to the five primary objectives above are also posed. These
questions are systematically addressed throughout the report with recommended resolutions
provided in Chapter 4.
• [Fatigue loading – Articles 3.6.1.4, 6.6.1.2.1] Is the current fatigue load model in terms of truck
position (i.e., single design truck passages positioned in various transverse lane positions)
appropriate for cross-frame analysis and design or should multiple presence effects be
considered?
• [Fatigue loading – Article 3.4.1] Are the current AASHTO LRFD Fatigue I and II load factors,
which were calibrated for girder force effects and recent WIM data, appropriate for cross-
frame analysis and design?
• [Fatigue loading – Table 6.6.1.2.5-2] Is the “n” value (i.e., number of cycles per truck
passage) currently assumed for the generic “transverse member” designation applicable for
cross-frames?
• [Fatigue loading – Article 6.7.4.2] What is the influence of composite bridge geometry
(e.g., support skew, horizontal curvature) and cross-frame layout (e.g., cross-frame spacing,
staggered layout) on the load-induced fatigue behavior of cross-frame systems (including
governing force effects, critical lane loading, and critical members under AASHTO fatigue
loading criteria)?
• [Fatigue loading – Articles 6.7.4.1, 6.7.4.2] Based on the findings of the item above, is it neces-
sary to perform a refined analysis, using either simplified 2D or more advanced 3D methods
for straight and non-skewed bridges? More succinctly, are the minimum design requirements
outlined in Article 6.7.4.1 appropriate?
• [Analysis – Article 4.6.3.3.4] Is the current established R-factor (0.65AE), which was based on
analytical and experimental studies of a noncomposite system, appropriate for cross-frames
in the composite condition or should alternative 3D modeling approaches be developed for
cross-frames?
• [Analysis – Article 4.6.3.3.2] Expanding on White et al. (2012), what are the limitations of
simplified 2D analysis techniques commonly used by popular commercial bridge software
programs in terms of predicting cross-frame force effects in composite systems? Are there
methods available to improve these simplified techniques?

• [Stability bracing – Article N/A] Can the AISC design guidelines for stability bracing be
incorporated into AASHTO LRFD? Are special requirements needed for negative moment
regions of continuous systems?
• [Stability bracing – Article N/A] How are these stability bracing requirements combined with
other load conditions such as wind?
1.3 Scope of Study

Objectives (a) through (e) and the specific questions identified in the preceding subsection
cover a wide range of behaviors and functions of cross-frames. Not only are the research
objectives comprehensive in nature, but the term “cross-frame” encompasses a variety of
systems and conditions as well. As noted previously, cross-frame systems are used in many steel
bridge applications (e.g., I-girder bridges versus tub-girder bridges, intermediate cross-frames
versus cross-frames at supports) throughout the entire construction and service life. As such,
their behavior must be evaluated in both noncomposite and composite systems. In the context
of tub-girder bridges, cross-frames internal to the tub girder primarily control distortion of
the shape and will generally have no fatigue issues, nor do these braces significantly enhance
the stability of the tub girder. Although studies related to fatigue will have application to the
behavior of external cross-frames in tub-girder systems, these girder systems were not con-
sidered in this study.
Additionally, cross-frame configurations and details come in many shapes and sizes. Specific
design and connection details are typically recommended on a state-by-state basis in the United
States. Each state Department of Transportation (DOT) (or the equivalent governing agency)
regulates and unifies bridge design in that state through standard details, specifications, and
design guidelines. For instance, cross-frame members are commonly comprised of single-angle
sections, but WT and double-angle sections are also occasionally used. Similarly, cross-frames
are generally configured as X- or K-type, but different variations are utilized throughout the
country. State standards also differ in how cross-frame members are connected to gusset and
connection plates (i.e., welded versus bolted). An industry survey was conducted as a part of the
investigation and is outlined in Section 2.1 to summarize and compile all of these unique details.
To maintain a manageable scope, the research team focused its efforts on the most common
and practical conditions found in steel I-girder bridges. As such, the research conducted and
documented herein is dedicated only to X- and K-type intermediate cross-frames comprised of
single-angle sections in steel I-girder highway bridges. As is discussed throughout the report, the
effects of different angle sizes and connection details on cross-frame performance are examined,
as are parameters related to the bridge geometry (e.g., girder spacing, deck thickness, etc.).
1.4 Organization of Report

Including this background section, this report is organized into four chapters, followed
by six appendices. Chapter 2 outlines the research approaches adopted to address the major
objectives of NCHRP Project 12-113. More specifically, the flow of work that includes field
experiments, model validation, and analytical studies is detailed. Four different analytical
studies are described, including the major modeling assumptions and the specific parameters
evaluated. Chapter 3 subsequently presents summarized results of the experimental and analytical
studies. For each major study, key observations and findings are outlined. Chapter 4 then
synthesizes the key findings from Chapter 3 in the context of the AASHTO LRFD provisions
identified previously in Section 1.2. Specific suggestions for implementation of the findings are
provided, as are areas where further research would be valuable.

Background 11
Chapters 2 and 3 of the main report are not intended to be comprehensive in nature. Rather
than overwhelm the reader with the full set of experimental and analytical data, sample results
are selectively presented to provide the proper context to the major conclusions summarized in
Chapter 4. For a more detailed overview of the experimental and analytical studies and results,
several appendices have been prepared.
Based on the key findings summarized in Chapter 4, Appendix A includes the draft speci-
fication revisions prepared for consideration by the AASHTO T-14 Technical Committee for
Structural Steel Design and the AASHTO Committee on Bridges and Structures. Appendices B
and C present cross-frame design examples for a straight bridge with normal supports and a
horizontally curved bridge with radial supports, respectively. These examples articulate how
the various load cases and conditions independently investigated in Chapters 2 and 3 can be
combined for the various stages of construction and service. Rather than develop representative
examples from scratch, the research team elected to adopt and modify previously published
design examples that are commonly referenced by design engineers. As such, the design
examples generally follow Steel Bridge Design Handbook (SBDH) Examples 2A (Barth 2015)
and 3 (Rivera and Chavel 2015).
Appendices D through F cover much of the same material as in the main body of the report.
However, these appendices provide a more thorough and exhaustive overview of the research
methodology, results, and critical findings. Appendix D specifically addresses Phase I of NCHRP
Project 12-113, whereas Appendices E and F address Phases II and III, respectively. The various
phases of the projects are detailed in Chapter 2.

CHAPTER 2
Research Approach
To address the major objectives of NCHRP Project 12-113, four distinct phases were system-
atically carried out by the research team. The four phases, as identified in the project RFP, are as
follows: (I) Planning, (II) Part 1 of the Analytical Program and Field Experiment, (III) Part 2 of
the Analytical Program and Proposed AASHTO Specifications, and (IV) Final Products. Each
successive phase built on the findings and lessons learned from the previous phase. This is
demonstrated schematically with a workflow diagram in Figure 2-1. As noted in Section 1.4,
comprehensive summaries of Phases I, II, and III are provided in the respective Appendices D, E,
and F. The information provided in this chapter is intended only to provide a general overview
of the research approach and methodologies.
In Phase I, an extensive literature review and industry survey was conducted, and preparations
were made for the Phase II field experiments. Phase I consisted of six tasks, which are briefly
summarized below:
• Task 1: Conduct a literature review, document current design practices, and examine how
design software incorporates these design practices.
• Task 2: Synthesize the results of the literature review to identify the knowledge gaps related
to the research objectives.
• Task 3: Propose an analytical program to be executed in two parts. Part 1 (executed in
Phase II) includes preliminary analysis to aid in the development of the experimental
program, and Part 2 (executed in Phase III) includes comprehensive parametric studies to
achieve the research objectives outlined in Chapter 1.
• Task 4: Propose field experiments (executed in Phase II) that include one straight bridge,
one skewed bridge, and one horizontally curved bridge. Measurements include (i) cross-
frame member fatigue force ranges under controlled application of live load and (ii) in-service
effective and maximum stress ranges for the same cross-frame members.
• Task 5: Identify existing articles of AASHTO LRFD that require modification.
• Task 6: Prepare Interim Report No. 1.
Note that several of these tasks were previously summarized in Chapter 1, including the
review of pertinent research and AASHTO LRFD articles. Section 2.1 briefly details the industry
survey performed by the research team that facilitated the planning of the experimental and
analytical studies and also aided in the development of the research scope. Appendix D also
outlines this material in a more thorough manner. The experimental and analytical programs
are covered under the Phase II and III tasks discussed in the following paragraphs.
In Phase II, field experimental studies and model validation were performed, which allowed
the analytical studies in Phase III to commence. In total, Phase II consisted of four tasks, which
are briefly summarized below:
12

Research Approach 13
Phase Perform background

I review and industry survey
Phase Conduct field experiments

II and obtain measured data
Phase Validate FEA modeling

II approach
Phase Conduct parametric FEA

III studies
Phase Develop specification and

III commentary language
Phase Finalize report and prepare

IV AASHTO ballot items
Figure 2-1. General flow of work related to NCHRP Project 12-113

(FEA = finite element analysis).
• Task 7: Execute Part 1 of analytical program related to the specific bridges evaluated in the
field experiments.
• Task 8: Execute the field experiments, as introduced in Task 4.
• Task 9: Validate the analytical modeling approach based on the results of the field experiments.
The methodology used to accomplish Tasks 7 through 9 is briefly outlined in Section 2.2.
Appendix E provides a more detailed overview of these Phase II tasks.
Phase III expanded on the work completed in Phase II. In order to properly evaluate the
behavior and design criteria of cross-frames in common highway I-girder bridge systems,
a much larger representative sample than the three unique bridge geometries and traffic condi-
tions from the field studies was necessary. As such, the validated modeling approach established
in Phase II was used to conduct extensive analytical studies that ultimately improve the breadth
and depth of knowledge pertaining to cross-frame behavior. Phase III included three major
tasks, as summarized with the following:
• Task 11: Execute Part 2 of the analytical program. Perform parametric studies related to live
load-induced fatigue performance and stability bracing characteristics of cross-frames.
• Task 12: Based on the analytical and experimental investigations, develop specification and
commentary language for proposed changes to AASHTO LRFD.
Sections 2.3 through 2.6 describe the methodology used to carry out Task 11 and the analy
tical studies. The project culminated in Phase IV, for which the proposed specification and
commentary language was finalized for potential balloting by AASHTO. These proposed modi-
fications are summarized in Chapter 4 and provided in full in Appendix A.
As outlined in Section 1.2, there were five major objectives of the NCHRP Project 12-113
investigation. Task 11 was structured in such a way to address these objectives through a series

Table 2-1. Overview of Task 11 work breakdown.
Objective Overview Model Software
a Fatigue loading studies

Fatigue Loading Study Abaqus
Influence of bridge geometry and
b
cross-frame details
c Cross-frame stiffness reduction R-Factor Study Abaqus
Commercial Design Software A,

d Commercial software evaluation
Software Study Software B
e Stability bracing requirements Stability Study Abaqus
of four independent computational studies. Table 2-1 summarizes these independent studies

as well as the various software platforms used.
The studies related to Objectives (a) and (b) in Table 2-1 were performed together via
a parametric study in Abaqus and are abbreviated as the “Fatigue Loading Study” herein.
The Fatigue Loading Study addresses the first of the three major research focuses: (i) fatigue
loading criteria. Objective (c), related to cross-frame stiffness and eccentric connections, was
also accomplished in a standalone parametric study in Abaqus/CAE (2017). This is referred to as
the “R-Factor Study.” The studies related to Objective (d), abbreviated as the Commercial
Design Software Study,” draw comparisons between commercially available software programs
(Software A and Software B) and validated models in Abaqus. Both the R-Factor Study and the
Commercial Design Software Study address the second of the three major research focuses:
(ii) analysis techniques.
Before addressing Objective (e), there are two important notes to make with respect to
the Commercial Design Software Study. First, no references to the specific commercial software
packages are made throughout this report to avoid critiquing or endorsing a specific brand.
As such, the two software packages studied extensively are generically referred to as Software
A and B. Second, the term “commercial software program” must be clearly defined. The term
“commercial” can refer to any program available to the public with a one-time or annual fee.
However, the focus of the Commercial Design Software Study is on those commercially avail-
able programs specifically designed for and marketed as a bridge design tool. Abaqus, although
a commercial program, is a general purpose finite element analysis (FEA) software program
and is not specifically suited for bridge design, whereas Software A and B are. Abaqus is one
of many general purpose FEA programs commonly used in research. As noted in Section 2.2,
the modeling approaches in Abaqus were adjusted based upon validation data from the field
monitoring. With that in mind, the research team could have also used a variety of other general
purpose FEA packages and completed the same validation.
The study related to Objective (e), which involves the stability bracing component to the
project, was performed independently of the others and is identified as the “Stability Study” in
this report. The Stability Study investigates the stiffness and strength requirements for cross-
frames, which act as torsional braces for I-girders. This study addresses the final of the three
major research focuses: (iii) stability bracing requirements.
The methodology used for the Fatigue Loading Study is summarized in Section 2.3, whereas
the methodology for the R-Factor, Commercial Design Software, and Stability Studies are
summarized in Sections 2.4, 2.5, and 2.6, respectively.

2.1 Industry Survey

An industry survey was developed and distributed by the research team to gather informa-
tion from DOTs and consultants in the following areas: (i) commonly used software packages,
(ii) fatigue design of cross-frames, (iii) load-induced fatigue issues with cross-frames encoun-
tered in existing bridges, and (iv) standard cross-frame details. In total, there were 57 responses
to the distributed survey representing a variety of states and consultants. Given the high response
rate, especially from DOTs, the results of the survey served as a reliable tool with respect to
focusing research efforts in the remaining phases of the project.
This section summarizes many of the key observations obtained from the industry survey,
which further emphasize the research motivations outlined in Section 1.1. For a detailed review
of the survey and responses, refer to Appendix D.
In terms of commercial design software use and fatigue design procedures, the following
observations were made:
• A wide range of commercial design software packages are used for bridge design in the
United States. Software B (2D program) was the most popular among bridge owners and
consultants, and Software A (3D program) received the third most votes. Given their wide-
spread use, Software A and B were selected by the research team for further investigation in
Phase II and III of the project.
• In many cases, respondents indicated that their software of choice (i.e., not limited to just
Software A and B) could not perform fatigue design checks or that they were unsure if it had
such capabilities.
• Several respondents indicated that they have encountered issues or had concerns related
to the fatigue checks made by their software of choice. This was largely attributed to the
“black box” nature of many software programs, as it was not transparent how the design
checks were performed.
• As a spot check for the analysis program, many respondents indicated that they also use
hand calculations to perform fatigue design checks for cross-frames.
In terms of load-induced fatigue issues encountered in existing bridges, the following
observations were made:
• Only 28% of the respondents indicated that their organization has previously experienced
load-induced fatigue cracking in existing bridges and cross-frames.
• In many of those cases, the respondents noted that load-induced fatigue often worked in
tandem with corrosion-related section loss and distortion-induced fatigue.
• Multiple responses pertaining to distortion-induced fatigue were recorded. However, as noted
in Chapter 1, the focus of the present study is on load-induced fatigue effects.
In terms of standard cross-frame details used across the country, the following observations
were made:
• Since many states provide multiple standard details depending on bridge geometry and
force demands, a total of 58 details (including cross-frames at supports) were compiled and
reviewed. The following characteristics were of interest when reviewing typical cross-frame
details: cross-frame configuration/layout (X-frame, K-frame, or Z-frame), member sections,
member end connections (bolted or welded), and use of gusset plates.
• Single-angle sections are the most common structural shape specified for the cross-frame
members, ranging from L3×3×5/16 to L6×6×5/8. In some instances, DOTs do not provide
typical sizes. For these cases, it is the responsibility of the designer to conduct a rational
analysis and determine the required size.

(a) X-Type (b) X-Type; No Top Strut (c) X-Type; No Struts (d) K-Type
(e) K-Type; No Top Strut (f) Inverted K-Type (g) Z -Type
Figure 2-2. Various cross-frame configurations used across the United States.
• Cross-frame connection details vary substantially from state to state. Figure 2-2 graphically
summarizes the range of cross-frame configurations used in the United States.
• In almost all instances where gusset plates are used to attach cross-frame members to
the connection plates, the gusset plates are bolted to the connection plates and welded to the
member. The lone exception to this is the Texas DOT (TxDOT) cross-frame details, where
cross-frames often utilize erection bolts and the final connections are welded prior to deck
construction.
• In instances where gusset plates are not used to connect members to the connection plates,
the diagonals are connected in one of two ways. They are either bolted directly to the connec-
tion plate or they are fillet welded to the top and bottom chords.
As indicated above, many DOTs specify bolted connections between the cross-frame member
and the gusset plate with TxDOT being a key exception. In these instances, slip-critical (SC)
connections are commonly used. Given that the present study is generally focused on the
cross-frame response within the elastic range (i.e., the fatigue limit state), it is assumed that
SC bolted connections do not slip at these service-level loads, such that the response of a cross-
frame is relatively unaffected by the connection method. Therefore, the studies presented herein
do not differentiate between bolted and welded connections, as cross-frame behavior is more
influenced by its member size and overall geometry.
As noted in Section 1.3, the industry survey allowed the research team to develop a manageable
scope for the analytical studies. Based on these results, only single-angle sections in traditional
X-type and K-type cross-frames were assessed [i.e., cross-frames (a) and (d) in Figure 2-2].
It should also be noted that cross-frames without top and/or bottom struts can potentially lead
to stability issues during construction. Although zero-force members in many stability applica-
tions (i.e., construction condition), struts are important since they prevent buckling mode in
which the girders rotate out-of-plane in opposite directions. Thus, it is always recommended to
include top and bottom struts for both X- and K-type frames, particularly in horizontally curved
bridges and/or bridges with significant skews.
2.2 Field Experimental Program and Model Validation

For Phase II of the project, three different bridges were instrumented with strain gages and
experimentally tested with two different methods: (i) a controlled live load test and (ii) in-service
monitoring of effective and maximum stress ranges. As noted previously, three different bridge
geometries were explored as part of this field study, including a straight bridge with normal

supports, a straight bridge with skewed supports, and a horizontally curved bridge. The research
team assessed the pros and cons for a variety of existing bridge structures throughout Texas
but ultimately elected to study three in the greater Houston area.
This section provides a brief overview of the field experimental program and the general
procedures used to validate the FEA modeling approach. Section 2.2.1 succinctly describes the
key geometric parameters and traffic conditions of each instrumented bridge. Section 2.2.2
outlines the two experimental tests performed with additional guidance on how the data was
processed. Note that the commentary provided herein is not intended to cover every aspect of
the field tests. For a more detailed synopsis, refer to Appendix E.
2.2.1 Instrumented Bridge Overview

The straight bridge with normal supports (referred to as instrumented Bridge 1 hereafter)
serves as an off-ramp for traffic traveling northbound (NB) on IH-45. The bridge contains
13 spans, including 10 spans of prestressed concrete girders and a three-span continuous unit
of built-up steel plate girders. In the steel spans, five nonprismatic girders across the width
support three lanes of traffic. The respective lengths of the steel I-girder spans are 194 feet,
240 feet, and 194 feet. X-type cross-frames are oriented normal to the longitudinal axis of the
bridge at a 19-foot typical spacing. According to the 2007 construction drawings, the estimated
average daily traffic (ADT) for this bridge is approximately 3,300. Only the southernmost end
span (194-foot) was instrumented with strain gages due to favorable ground access in the field.
The skewed bridge (referred to as instrumented Bridge 2 hereafter) serves five lanes of south-
bound (SB) IH-45 traffic, which is a major arterial road through Houston. Instrumented Bridge
2 is a three-span continuous, steel I-girder bridge with a total length of 465 feet (125-foot,
215-foot, and 125-foot spans). A 42-degree support skew relative to the centerline of the bridge
was used to accommodate the roadway below. The five lanes of traffic are supported by 12 non-
prismatic steel I-girders; thus, the superstructure is highly redundant. X-type cross-frames are
oriented normal to the longitudinal axis of the bridge at a 17-foot typical spacing. Near the
skewed supports, the designers selectively eliminated cross-frame panels in accordance with
AASHTO LRFD Article 6.7.4.2 guidance. In these cases, lean-on braces were used to mitigate
load-induced force effects. According to Texas DOT traffic maps (2020), the ADT for this
bridge is 120,510, which provided excellent loading conditions in terms of the in-service moni-
toring study. Only the northernmost end span (125-foot) was instrumented with strain gages
due to favorable ground access and traffic control requirements.
The horizontally curved bridge (referred to as instrumented Bridge 3 hereafter) is a direct
connector along the SH-146 corridor that primarily serves large trucks traveling to nearby
shipping ports. The bridge contains 14 spans, including 10 straight spans of prestressed concrete
girders and a four-span continuous unit of curved steel I-girders. The radius of curvature of
instrumented Bridge 3 is 800 feet, and the supports are normal to the centerline of the bridge.
The respective lengths of the steel spans are 255 feet, 310 feet, 238 feet, and 216 feet. The four
girders across the width support one lane of striped traffic. X-type cross-frames are oriented
normal to the longitudinal axis of the bridge at a 12.5-foot typical spacing. According to the
2009 construction drawings, the ADT is approximately 40,000. Only the 216-foot end span was
instrumented with strain gages due to favorable ground access. For additional information on
the instrumented bridge geometries and cross-frame details, refer to Appendix E.
2.2.2 Experimental Testing Program Overview

For each bridge, several cross-frame members and the bottom flanges of neighboring
girders were selected for strain gage instrumentation. The locations of these gages were based

on numerous factors including field access, traffic control considerations, the capacity of the
data acquisition system (DAQ), and preliminary analysis. The preliminary analysis, which was
conducted prior to the field studies, identified the cross-frames likely to experience the largest
stress ranges, as well as the critical truck positions to evaluate in the controlled live load test. The
results of these preliminary studies are further detailed in Appendix E.
A wireless monitoring system was designed to collect continuous strain time-histories, as well
as simultaneously calculate stress cycle counts in accordance with commonly used rainflow-
counting algorithms (Downing and Socie 1982). For each instrumented cross-frame member,
four quarter-bridge strain gages were installed at various locations along a cross-section. By
measuring the strain at various points on a cross-section, the axial component of stress could
be isolated from the biaxial bending effects in the member via the linear regression method
developed by Helwig and Fan (2000). This procedure was validated by laboratory studies prior
to implementation in the field.
Given that the Category E′ designation established for these welded connection details was
explicitly based on axial stresses (i.e., P/A), reliably measuring axial stress was critical. It
is important to note, however, that significant bending stresses were measured as part of this
study to verify the impact of eccentric end connections. Two strain gages were also installed at
each edge of the girder bottom flange, such that longitudinal bending stress and warping stress
components could be independently evaluated.
Upon completing the strain gage installation, two different experimental tests were performed
for each bridge. The controlled live load test consisted of four dump trucks loaded with sand
(of known axle configuration and weight) positioned at various longitudinal and transverse
points on the concrete deck. The in-service monitoring consisted of recording live load stress
range cycles in the instrumented elements over a one-month period. The subsections herein
briefly address each test separately.
2.2.2.1 Controlled Live Load Test

In total, seven different static load cases were considered for the controlled live load test,
as well as a set of pseudo-static moving loads (i.e., trucks driven at slow speeds) to generate
influence lines for each instrumented component. Cross-frame stresses, girder flange stresses,
and select girder displacements were measured during each successive load case. Refer to
Appendix E for full instrumentation plan and results, including sketches of strain gage locations
and static load cases performed on each bridge.
As noted in Chapter 1, the accuracy of even the most sophisticated FEA analysis is limited
by the modeling assumptions. Thus, the field-measured data obtained from the controlled live
load tests were vital in validating the FEA models used in the Phase II parametric studies. The
load cases from the field studies were effectively replicated in a 3D FEA model, which allowed
the research team to fine-tune the parameters and assumptions to achieve a close match of the
actual conditions of the bridge system. The modifications required to obtain good agreement
with the measured data are detailed in Section 3.1. Parameters such as boundary conditions
and the presence of concrete barriers were explored.
2.2.2.2 In-Service Monitoring

For the in-service monitoring phase of the field tests, continuous strain history over a four-
week period was converted into histograms of stress cycle counts (i.e., a spectrum of stress range
measurements). Two key performance characteristics of the instrumented cross-frame members
and girder flanges can be inferred from these spectra: (i) the number of truck events causing a
significant stress cycle and (ii) the stress cycle magnitudes that a structural component typically

experiences. Note that the one-month monitoring period is sufficiently long to capture stabi-
lized data as well as hourly, daily, and weekly trends that may deviate from normal or average
conditions (Connor and Fisher 2006; Fasl 2013).
In addition, three supplementary metrics derived from the histogram plots were used by the
research team to compare data between the various instrumented components of the different
bridges. These three metrics are effective stress range (Sre), maximum stress range (Srm), and
index stress range (Sri).
All three, which can be derived from simple postprocessing of the measured data, are
effective tools for evaluating the fatigue performance of structural components and details.
Effective and maximum stress ranges are defined in the current 9th Edition AASHTO LRFD
Specifications (2020). Index stress range was a metric developed by Fasl (2013) but is not
currently adopted by standard codes. These metrics are further explained in the context of
measured data in Section 3.1 and Appendix E.
2.3 Fatigue Loading Study

The Fatigue Loading Study is intended to investigate two major questions concerning
load-induced fatigue design of cross-frame systems: (i) the influence of bridge geometry
(e.g., support skew, horizontal curvature) and cross-frame details (e.g., cross-frame layout) on
cross-frame fatigue force effects and (ii) the appropriate fatigue stress ranges for the evaluation
of cross-frames in right, skewed, and horizontally curved bridges. These questions represent
Objectives (a) and (b) established in Section 1.2.
Preliminary answers to these questions were obtained from the field experiments conducted
during Phase II. However, the instrumented bridges represent only three unique framing
geometries subjected to local traffic conditions in the greater Houston area. In order to
adequately assess the AASHTO fatigue loading criteria, a more expansive study was required.
Based on the lessons learned from the modeling validation process in Phase II, the research
team developed and conducted a robust, finite element parametric study. The parametric study
effectively improved the depth of knowledge by expanding the breadth of the data set. In total,
4,104 unique bridge geometries were studied including various girder and cross-frame layouts,
girder cross-sections, and cross-frame details. The 4,104-model data set is intended to reason-
ably represent a wide variety of steel I-girder bridges currently in service in the United States. The
details of the 4,104-model matrix are briefly discussed in the subsequent sections.
To address the major objectives of the Fatigue Loading Study outlined above, the research
team focused its efforts on three major tasks. First, a unified modeling approach was developed
to analyze the various bridge structures (i.e., right, skewed, and horizontally curved bridges)
in a consistent and repeatable manner. Second, the research team implemented the current
fatigue design criteria on the full set of bridges. The cross-frames of the representative bridges
were effectively designed for AASHTO LRFD 9th Edition fatigue provisions. Lastly, measured
high-resolution WIM data from different U.S. states were implemented on a subset of the
bridges considered in the study.
The observations and findings of each successive task were used to address Objectives (a)
and (b) outlined in Section 1.2. By comparing the load-induced force response of cross-frames
in a variety of bridge types, the influence of bridge geometry and cross-frame details can be
evaluated. By then comparing the fatigue design forces with the force effects due to real traffic
patterns (i.e., via WIM records), a realistic approximation of actual fatigue stress ranges can be
assessed. The appropriateness of current AASHTO LRFD fatigue design criteria such as critical

load position and Fatigue I and II load factors [which have been calibrated for longitudinal
girders only through the research efforts documented in SHRP 2 R19B (Modjeski and Masters
2015)] can also be evaluated with respect to cross-frame systems.
Following this introduction, the modeling approach adopted in the Fatigue Loading Study is
outlined (Section 2.3.1). The procedures used to evaluate the bridges for current fatigue design
loading are presented (Section 2.3.2), as well as the procedures related to the WIM study (Sec-
tion 2.3.3). Major results and outcomes related to the studies are summarized in Section 3.3.
2.3.1 General Methodology

Preliminary work and preparation are critical for any comprehensive analytical parametric
study. It was especially important for the Fatigue Loading Study, which parametrically consid-
ered entire bridge structures in refined 3D analyses. As such, the research team dedicated its
efforts in the early stages of Phase III to prepare for and expedite the large-scale analytical
study. Due to the wide range of parameters considered, an ordered process that systematically
conducts the analyses and processes the results was developed. The balance between compu-
tational efforts and modeling accuracy was of particular interest.
For a parametric study of this size, it was not feasible to evaluate every permutation of every
critical parameter given limitations to computational resources and time. With computational
constraints in the forefront of the discussion, the research team conducted preliminary analyses
to establish expectations and develop a reasonable scope. In particular, the seven tasks below
were performed prior to conducting the Fatigue Loading Study to help narrow the focus and
limit the required computational time:
• Adopt a consistent modeling technique for cross-frame elements;
• Develop a reasonable analytical testing matrix for purposes of the Fatigue Loading Study;
• Refine the applied loading approach;
• Conduct mesh sensitivity studies to optimize computational efforts;
• Streamline postprocessing efforts by evaluating only critical cross-frames;
• Develop scripts to automate the modeling process; and
• Develop Excel-based macros for processing the FEA results.
Each of these preliminary tasks was important in developing a robust study. For purposes of
this report, each task is cursorily addressed herein with the following subsections; for a more
detailed overview of the methodology used, refer to Appendix F.
2.3.1.1 Modeling Assumptions

As documented in subsequent sections of the report, there are numerous ways that designers
commonly model bridge superstructures, ranging from sophisticated (3D) to simplified (2D)
approaches. For each approach, there are several widely regarded methods specifically directed
at modeling cross-frames. This section does not address and compare every modeling tech-
nique. Rather, a unified modeling approach that is both repeatable and justifiable is explored
for use in the Fatigue Loading Study. For additional information, Sections 2.4 and 2.5 provide
an overview on the most common modeling techniques.
Because they generally produce more reliable cross-frame results, only 3D models were
considered for use in the Fatigue Loading Study. Most 3D models, either from commercial
design software packages or high-fidelity FEA software, employ the same basic modeling
strategies for the concrete deck and girders. The strategies for cross-frames in 3D models,
though, often vary. While the most rigorous and accurate representation of a cross-frame makes
use of shell elements, most common 3D models do not explicitly model the cross-frames with

shells but instead use line elements. While the present study utilized line elements with R-factors,
additional studies were also carried out making use of shell-element models. A discussion of
these modeling techniques is provided in the following paragraphs.
In terms of the more rigorous approach, each component of the cross-frame panel is
explicitly modeled with a shell-element: connection plates, gusset plates, cross-frame members
(typically single angles), and fill plates commonly found at the intersection of diagonals in
X-frames. The eccentric load path is also explicitly considered. These models are simply
referred to as shell-element models herein. Although generally more accurate, this refined
technique (especially with a fine mesh) rapidly increases the computational demands and
run time.
The second alternative, which is more common in design practice and even in many research
studies, simplifies the cross-frames as pin-ended truss elements. Therefore, the eccentric load
path and the stiffness contributions of connection and gusset plates are neglected. The actual
stiffness of the cross-frame system, though, can be approximated with a simple modifica-
tion factor (R-factor) assigned to the cross-sectional area of the truss element or the elastic
modulus of the material. These types of models are referred to as truss-element models in
this section.
Given the scope of the study, the shell-element modeling approach was deemed too
computationally intensive. Thus, the research team elected to perform the Fatigue Loading
Study with truss-element models to balance computational speed and modeling accuracy. To
better understand the behavior, preliminary studies were performed using the general purpose
FEA program, Abaqus, to determine which variation of truss-element models produces the
most accurate results relative to the validated, shell-element models. Based on these preliminary
studies, the following observations were made, which influenced the approach adopted for the
Fatigue Loading Study (Figure 2-3):
• Explicitly modeling the offset dimension between cross-frame working line and the girder
flanges better represents the stiffness of the panel and the inclination angle of the diagonal
members (as opposed to connecting the cross-frame member into a shared node at the
web-flange juncture);
• Without the connection plates, cross-frame forces are greatly affected by web distortion caused
by the concentrated forces acting on an unstiffened, flexible web (note that distortion effects
would not be impactful if cross-frames were connected directly into the girder flange); and
• Terminating the cross-frame truss member at the edge of or in the middle of the connection
plate affected the force distribution in the cross-frame members.
Figure 2-3. Final selected modeling technique

for truss-element cross-frames.

Note that the vertical offset dimension and the presence of connection plates are further
discussed in Sections 2.4 and 2.5, which focus heavily on the approaches traditionally adopted
by 3D commercial design software programs. Prior to conducting the parametric study, the
accuracy of the assumed truss-element modeling approach and its simplifications outlined
above was validated with the field-measured data. This procedure is detailed in Appendix F.
Lastly, it was determined that a first-order, elastic analysis was suitable for the fatigue load-
ing conditions that are representative of service-level load magnitudes. This approach is also
consistent with common design practice and AASHTO LRFD recommendations. To verify this
assumption, supplementary FEA studies that considered second-order effects and initial
imperfections in cross-frame members were performed, and their influence on the cross-frame
force demands was found to be negligible.
2.3.1.2 Analytical Testing Matrix

As previously stated, the bridges considered in the study were intended to represent a
common range of steel I-girder bridges currently in service. Developing a testing matrix that
represents thousands of steel I-girder bridges in the United States is challenging, considering
there are many different parameters that potentially impact the behavior and response of
cross-frames.
To establish a reasonable scope, 13 parameters were identified as independent variables (i.e., the
parameters to be evaluated throughout the study), and the remaining parameters were considered
constants. Table 2-2 describes the range of values used for the 13 independent variables. Note
that this list assumes that the truss-element modeling approach described in the previous sub-
section is implemented. Thus, parameters related to cross-frame details such as gusset plate
thickness are not considered in the Fatigue Loading Study but are instead considered in the
R-Factor Study. To facilitate the understanding of these parameters and the bridge geometries
evaluated, several sketches are provided in Appendix F (e.g., a sketch demonstrating how stag-
gered cross-frame layouts were considered).
Based on the parameters outlined in Table 2-2, it is apparent that conducting every permu-
tation of every parameter was not feasible given computational constraints. Thus, the matrix
Table 2-2. Range of values used for independent variables

in Fatigue Loading Study (L/d = span-to-depth ratio,
ksi = kips per square inch).
Parameter Range of Values

Number of spans {1, 2, 3}
L/d ratio {25, 30, 35}
Girder spacing [ft] {6, 8, 10}
Number of girders {3, 5, 7}
{0, 30, 60};
Support skew
{Parallel, Trapezoidal}
Radius of curvature [ft] {Infinite, 1,500, 750}
Cross-frame spacing [ft] {20, 30}
Cross-frame layout {Contiguous, Staggered}
Web depth [in] {72, 96}
Deck thickness [in] {8, 10}
Cross-frame type {X, K}
{2.86, 4.79};
Cross-frame area [in2]
{L4x4x3/8, L5x5x1/2}
Concrete modulus [ksi] {3,600, 5,000}

was filtered to generally provide only the cases that are likely most critical. Rules were estab-
lished to eliminate permutations deemed unnecessary or less important than others. Many
of these rules were based on current AASHTO provisions related to geometric limits. For
example, AASHTO Article 2.5.2.6.3 recommendations were implemented to establish reason-
able bounds on the span-to-depth-ratios considered in the study. To further clarify Table 2-2,
the following bulleted items provide key discussion points. The following list represents an
abbreviated version; refer to Appendix F for a more thorough overview and a comprehensive
list of every bridge parameter studied.
• The differentiation between independent variables and constants is based on prioritization
of the parameters, the information gained from the validation studies, data collected from
the industry survey, and common bridge design practice. For example, the presence of a
concrete barrier was deemed a pertinent variable based on the validation studies in Phase II,
as is documented in Section 3.1.1. However, for purposes of the Phase III parametric study,
the research team elected to simply maintain constant dimensions and parameters for barriers
rather than evaluate variable dimensions.
• In the same vein, overhang dimensions were taken as a uniform 3 feet (measured from
the centerline of the fascia girder to the outer edge of the deck) to simplify the calculations.
For bridges with a wider girder spacing, overhang lengths are commonly greater than 3 feet.
As is discussed later in the report, fatigue forces in cross-frames are often maximized by
truck passages where a wheel line rides along the overhang portion of the deck. It was found,
however, that force effects in a cross-frame member due to an “overhang” live load are gener-
ally linearly dependent on the outer wheel distance to the centerline of the fascia girders.
• Staggered cross-frame layouts were considered in the study, but skewed cross-frames were
not. For skewed bridges, cross-frames were either: (i) contiguous lines normal to the longitu-
dinal girders or (ii) discontinuous lines that run parallel to the support skew angle but placed
at a line normal to the longitudinal girders. Because both support skew angles considered in
the study (30 and 60 degrees) exceed 20 degrees, skewed cross-frames parallel to the support
skew were not considered in accordance with AASHTO LRFD Article 6.7.4.2.
• In general, the research team ensured AASHTO Article 6.7.4.2 was satisfied for the layout of
cross-frames. In particular, the cross-frame depth was selected to exceed 75% of the girder
depth to preclude any significant web-distortional effects. Additionally, the cross-frame
layout recommendations near obtuse and acute corners of skewed bridges given in AASHTO
Article C6.7.4.2 were followed.
• A constant stiffness modification factor (R-factor) is assumed for all cross-frame systems,
regardless of the assumed geometry and connection details. Because the focus of the Fatigue
Loading Study is not on stiffness modification factors, the research team elected to normalize
the effects of eccentric load paths for all cross-frame systems in the 4,104-model parametric
study. That way, any potential variability associated with the R-factors is eliminated, and
the relative impact is then consistent across all bridge models. A uniform factor of 0.60
was selected, as it was shown to be a reasonable representation of the “average” value in
preliminary R-factor studies.
• The decision to use R = 0.6 was made early in Phase III, several months before the final results
were obtained from the R-Factor Study (i.e., proposed R = 0.75). Clearly, consistently
larger cross-frame force effects would have been determined had the parametric study been
conducted with the larger R-factor. As noted in Section 3.3, an increase in R-factor from 0.60
to 0.75 would only increase those predicted force effects about 5% to 7%. As such, it was
justified that this deviation did not warrant a repeat of the parametric study since no major
conclusions outlined herein would be affected.
• The girder sections of each bridge were evaluated with respect to proportion limits (AASHTO
Article 6.10.2) and strength limit state resistance requirements (AASHTO Article 6.10.6).

In other words, it was ensured that the bridges studied represent realistic structures that have
been properly designed and detailed.
• Aside from the parameters listed in the tables, the research team also conducted spot checks
to examine the influence of several parameters including different flange dimensions and
the removal of bridge barriers.
2.3.1.3 Load Application

Current AASHTO fatigue loading criteria require that the design fatigue truck be positioned
in all transverse and longitudinal positions across the bridge deck to maximize the stress ranges
in the detail under consideration (Article 3.6.1.4.3). To implement these criteria, the research
team generated an influence-surface analysis in Abaqus by applying a unit point load across a
defined grid of load positions. The loading grid extended over the entire width (between the
inside faces of the barriers) and length of the bridge to capture the full influence. To then capture
the effects of a realistic truck passage instead of a static unit load, an Excel program and script
were developed to simulate these effects, as discussed in Section 2.3.1.6.
In terms of accuracy, it is important that the critical loading position is captured by the
specified loading grid. Preliminary studies were performed to optimize the loading grid dis-
cretization (i.e., how frequently is the unit load applied in the longitudinal and transverse
directions). Grids ranging from 1-foot to 5-foot longitudinal and transverse increments were
considered. From these analyses, it was determined that a relatively coarse discretization still
produced reasonable levels of accuracy in terms of predicting cross-frame force effects. The
applied loading grid was accordingly optimized by assigning load points at every 5 feet longitu-
dinally and at the following transverse locations: at each girder line, midpoint of every cross-
frame bay, and at a point 1 foot outboard from the centerline of each fascia girder.
2.3.1.4 Mesh Sensitivity

In addition to the sensitivity of the loading grid mesh, the sensitivity of the finite element
model mesh was also extensively studied. Fine element meshes potentially offer more accurate
solutions but at the cost of increased computation time and storage. Acknowledging that com-
putation efficiency was of the utmost importance for the Fatigue Loading Study, the research
team studied the effects of a coarser mesh on the predicted response of cross-frame members.
From preliminary studies, it was observed that cross-frame force effects were not overly
sensitive to the specified mesh size. Since localized stress concentrations or behaviors are not
important to the study at hand, a coarse mesh was adopted. The girder shells were meshed at
24 inches (i.e., three to four shells along the depth of the web), and the deck shells were meshed
at 48 inches (in addition to the loading grid). Quadrilateral meshes were typically used, except
at extreme cases near skewed ends. In those cases, special care was taken to ensure reasonable
aspect ratios (i.e., within 3 to 5).
2.3.1.5 Postprocessing Critical Cross-Frames

Rather than evaluating the forces in every cross-frame from every model in the Fatigue
Loading Study, it was more reasonable to evaluate only the most critical members. Knowing
that cross-frame force effects are likely dependent on load position, bridge geometry, and
location on the span, identifying the critical cross-frame for all 4,104 unique bridges was
challenging. Thus, the research team took a more systematic approach to predicting the most
likely “critical” cross-frames for any given bridge. Prior to the Fatigue Loading Study, a prelimi-
nary parametric study was conducted to gain a general understanding of (i) the typical location
of the critical cross-frame members and (ii) the influence of the transverse and longitudinal
load position on the force effects in these critical members.

Based on the findings of this preliminary study, the research team preselected cross-frames
to evaluate prior to running the full parametric study. Given the unique geometries and
cross-frame layouts, the selected cross-frames are not necessarily in the exact same location
for each bridge. Rather, general locations/regions that are consistent across all 4,104 models
were selected. These four general locations/regions, which apply equally to both straight and
horizontally curved bridges, included:
• Edge cross-frame bay closest to the point of maximum positive dead load moment (i.e.,
midspan for single-span bridges and approximately 0.35L measured from the end support in
end spans of continuous units);
• Interior cross-frame bay closest to the point of maximum positive dead load moment;
• Interior cross-frame bay nearest to the end support (skewed or non-skewed); and
• Interior cross-frame bay nearest to the intermediate support (skewed or non-skewed).
For these regions, every cross-frame member in the panel was examined (i.e., top strut,
bottom strut, and diagonals). These cross-frame panels were selected given the propensity
for more significant differential girder displacements in the surrounding regions. For a more
detailed evaluation of these assumptions, refer to Appendix F.
To illustrate these general locations, a representative sketch is provided in Figure 2-4. The
figure presents the critical cross-frame panels evaluated for representative single-span, two-
span continuous, and three-span continuous straight bridges. Note that these figures are simply
schematic and not drawn to scale. In addition, only intermediate cross-frames are illustrated;
cross-frames at end and intermediate supports are hidden for clarity, as these are not the focus
of the study.
2.3.1.6 Automation and Scripts

To expedite the analysis and postprocessing, the research team developed Python scripts
to automate the calculations. The Python scripts were designed to perform the following
Key: Critical cross-frame evaluated
Figure 2-4. Critical cross-frames evaluated for sample bridges:

(top) single-span with skewed supports, (middle) two-span
continuous with skewed supports, and (bottom) three-span
continuous with normal supports.

tasks: (i) develop the 3D FEA models for all geometries, (ii) run the simulated influence-
surface analysis, (iii) output the axial-force and displacement response of the preselected
cross-frame members of interest, and (iv) create summarized output files of all pertinent data
(i.e., influence-surface results for every critical cross-frame member). These extensive calcula-
tions were performed using the Texas Advanced Computing Center (TACC) supercomputers.
Once the summarized data files were obtained, further postprocessing was performed in an
external Excel program, which is documented in the next subsection.
2.3.1.7 Data Visualization

Due to the tremendous amount of data produced from the parametric study, the research
team identified a need to view and evaluate data in an organized manner. Considering that
4,104 bridges were analyzed, the total number of influence-surface results exceeded 65,000,
which is obviously not practical from a publication perspective. Thus, an Excel-based “data
visualization” workbook was developed as a centralized tool to view all influence-surface data
obtained from the finite element models.
An example showing two of the many influence-surface plots is presented in Figure 2-5.
These two plots correspond to a two-span, straight bridge with normal supports and an other
wise identical bridge (e.g., same deck thickness, girder spacing, etc.) with 60-degree support
skews. Thus, the only discernible differences between the bridges are the support skew and
the cross-frame layout near the skewed supports.
As depicted in the figure, influence-surface results are best displayed as color contour plots
where the x (longitudinal)- and y (transverse)-coordinates represent the spatial coordinates of
the applied load on the deck surface and the z-axis (color intensity) represents the corresponding
magnitude of the force effect. To clarify the contour data, a framing plan for each bridge
is also overlaid. Girder lines and cross-frames are depicted with thin green lines; supports
are represented as bolded and dashed black lines. The cross-frame of interest, for which the
influence-surface results are presented, are bolded. Note that the x- and y-coordinates are not
to scale; the transverse (y-axis) is scaled up for clearer viewing.
The contour plot in Figure 2-5 demonstrates the axial-force response of a diagonal cross-
frame member near the intermediate support. The diagonal that frames into the top flange of the
first interior girder from the left (i.e., “left” corresponding to negative values about the y-axis)
is displayed.
The color intensity on the plots corresponds to the axial force induced in the cross-frame
member due to a 1-kip load positioned at a given spatial coordinate on the deck surface. Thus,
the color intensity is specified in units of kips of axial force per kips of applied load on the deck
(i.e., kips/kip). For loads applied in the “red” area on the plot, the compression force in the
diagonal is maximized. For loads applied in the “blue” area, the tension force in the diagonal is
maximized. Load positions applied in the grey area have negligible influence on the axial-force
response in that cross-frame member. As an example, a value of +0.1 kips/kip at a given location
on the deck (representing a critical tensile load position) means that a 1-kip load applied there
results in 0.1 kips of axial tension in the cross-frame diagonal of interest.
From the sample results in Figure 2-5, it is obvious that the longitudinal load influence of the
skewed bridge is broadened relative to the straight bridge. In fact, there is strong tensile load
influence along the outer “right” edge of the bridge deck (i.e., a blue area), which is not present
for the straight bridge. The overall magnitudes at the critical load positions are also slightly
higher in the skewed system. For additional influence-surface results and commentary, refer to
Appendix F.

375
350
350
325
325
300
300
275
275
250
250
225
150 175 200 225

Longitudinal Position (ft)
150 175 200
125
125
100
100
75
75
50
50
25
25
-25 0
0
-13
13
-13
13
Transverse Position (ft) Transverse Position (ft)

0.2 (T)
-0.2 (C)
0.2 (T)
-0.2 (C)
Influence (kip/kip) Influence (kip/kip)
Figure 2-5. Comparison of influence-surface plots

for axial-force response of cross-frame diagonal
near intermediate support with (left) no skew and
(right) 60-degree skew (T = axial tension force,
C = axial compression force).
In total, the research team produced thousands of influence-surface plots similar to those in
Figure 2-5. Influence-surface plots are useful visual tools and provide important insights on the
load-induced response of cross-frames, particularly in the case of this study where evaluating
over 65,000 plots in an efficient and meaningful way is virtually impossible given the number
of parameters investigated and the variability in the response. While the contour graphs show
the response for a 1-kip force applied anywhere on the deck, the actual force generated in
the cross-frame due to a truck on the bridge is the sum of the wheel loads multiplied by the
corresponding influence-surface values at the wheel locations. In order to assess the correla-
tion between truck placement, fatigue stresses, and the various bridge geometries, additional
processing of the data was necessary. Thus, Sections 2.3.2 and 2.3.3 introduce realistic loading
conditions for these 4,104 bridge models.

More specifically, Section 2.3.2, utilizes the over 65,000 influence-surface plots to assess
current AASHTO fatigue loading criteria. Section 2.3.3 expands on that analysis and inves-
tigates the effects of measured WIM traffic streams.
2.3.2 AASHTO Design Loads

Although the unit loads outlined in Section 2.3.1.7 highlight key trends in the behavioral
response of cross-frame systems, the Fatigue Loading Study objectives can only be met by evalu-
ating the response under realistic traffic loads. This section focuses on the approach used for
current AASHTO design loads, whereas Section 2.3.3 focuses on the use of recent WIM records.
Results of these studies are subsequently presented in Chapter 3.
Implementing AASHTO fatigue design criteria involves an understanding of both the load
response and the corresponding resistance of a given detail. In terms of the organization of
this section, the methodology used to apply AASHTO fatigue loading on the influence-surface
results is first outlined. Load-induced fatigue resistance is covered later in the subsection.
To consider realistic traffic conditions on the numerous influence-surface results obtained,
an Excel program was developed to perform a series of bi-linear interpolation calculations.
Thus, truck passages were simulated over the various influence-surface plots to develop
influence-line plots, for which the axial-force response of a cross-frame due to the passing
truck is illustrated.
AASHTO Article 3.6.1.3.4 states that “a single design truck shall be positioned transversely
and longitudinally to maximize stress range at the detail under consideration, regardless of the
position of traffic or design lanes on the deck.” To satisfy this requirement, a single AASHTO
fatigue truck is analytically traversed along the entire length of the influence-surface in a speci-
fied lane (i.e., the transverse position was held fixed for the entire longitudinal passage, such
that “zig-zag” loads were not considered). Different lane passages were considered by repeating
the process in 1-foot transverse increments in both the forward and backward directions
(i.e., truck traversing upstation and downstation). At deck overhangs, a 1-foot clear distance
between the centerline of the outermost wheel line and the inside face of the barrier was
maintained. Although consideration of trucks driving in close proximity to the barrier (within
1 foot) is not required (particularly for fatigue evaluation) because it is an infrequent occurrence
in most cases, these extreme loading conditions were considered in the study for completeness.
An example of this procedure is presented in Figure 2-6. The sample influence-surface plot
corresponds to a cross-frame diagonal near the intermediate support of a heavily skewed bridge
(i.e., one of the 4,104 bridges studied). The AASHTO fatigue truck was analytically traversed
across the influence surface in the upstation direction for three distinct lane positions: left wheel
line 1-foot from the left barrier, centerline of bridge, and right wheel line 1-foot from the right
barrier. Note that in the actual analyses of this particular influence surface, this process was
completed many more times to consider all possible lane passages between the barriers.
The x-axis of the influence-line plots corresponds to the longitudinal position of the front
truck axle relative to the coordinate system established in the color contour plot. Therefore, at
skewed ends, the truck is introduced to the deck surface at different x-coordinates, as is reflected
in the figures.
As presented in Figure 2-6, the axial-force response of the cross-frame varies significantly for
each truck passage. When traversing the bridge along the left barrier, the cross-frame member
of interest experiences a force reversal, resulting in a substantial force cycle (11.8 kips). For the
fatigue truck passing along the right barrier, a large force reversal is also observed, except in
the opposite order (9.4 kips). For the truck passing along the centerline of the bridge, the force

Transverse Position (ft)

-17 0 17
10
Axial Force (kips)
420
11.8 kips
0
370
-5
-10
-30 20 70 120 170 220 270 320 370 420
320
Longitudinal Position of Truck (ft)
10
270
Axial Force (kips)
5
3.7 kips
220
0
-5 0.2 kips 0.5 kips
170
-10
-30 20 70 120 170 220 270 320 370 420
120
10
Axial Force (kips)
70
5
9.4 kips
0 20
-5
-10
-30
-30 20 70 120 170 220 270 320 370 420

Figure 2-6. Sample data showing the development of influence-line

plots from an influence surface.
magnitudes are smaller, but the response is more complex. Whereas the other two truck passages
resulted in one primary force cycle, the centerline passage results in additional secondary
cycles of lesser magnitude. The primary cycle (3.7 kips) and the two secondary cycles (0.5 kips
and 0.2 kips) are illustrated in the figure.
Given that the permanent stress states in the cross-frames of these representative bridges are
unknown (e.g., locked-in fit-up forces, dead loads, residual stresses), only live load force/stress
cycles entirely in tension or subject to reversal, regardless of how small the tension component
is, were considered. It is recognized that, in actuality, this assumption would only be valid when
the Fatigue I factored tensile stress component exceeds the compressive stress due to unfactored
permanent loads or the permanent loads are tensile in nature. To simplify the analysis, the
tensile force component was always considered to be large enough to propagate a crack.
Thus, truck passages that induce a purely compressive force/stress cycle with no tensile force
component are disregarded from the evaluation, as they are not a fatigue-sensitive loading
condition. Also note that, since the primary focus of this study is related to the fatigue limit

state, permanent (dead) loads and locked-in stresses are not explicitly addressed herein,
except as described above.
The computationally intensive procedure introduced in Figure 2-6 was performed for all
influence-surface plots obtained from the 4,104 bridge models analyzed in Abaqus, which
resulted in over 3 million influence-line plots similar to the three examples presented in the
figure. A script was developed to automate these moving-load simulations. The script was
designed to perform rainflow counting on the various influence-line plots (i.e., axial-force
time-histories), and the following output was recorded for each cross-frame member evaluated:
(i) the transverse position of the AASHTO fatigue truck that maximizes the force range, (ii) the
magnitude of the primary force cycle caused by the critical lane passage, and (iii) the number
and magnitude of any secondary cycles caused by that same critical lane passage.
The maximum force range for a given cross-frame member represents the unfactored design
force for which the engineer evaluates the Fatigue I or II limit state. Note that many of the results
presented in Chapter 3 are based on the unfactored design force ranges obtained from this
specified procedure.
The unfactored fatigue force ranges were then factored in accordance with current 9th Edition
AASHTO LRFD load factors (as well as the cross-frame-specific load factors proposed and
outlined in Section 3.2) and converted into an axial stress with consideration of shear lag effects
[i.e., the U factor specified in AASHTO Table 6.6.1.2.3-1 (Condition 7.2)]. Dynamic load
allowance (1.15 for fatigue limit state) was also subsequently applied to produce factored axial-
stress ranges for fatigue design and evaluation.
The resistance side of the AASHTO fatigue design criteria (i.e., load-induced fatigue) was then
introduced into the computational studies. Based on the guidance provided in Article 6.6.1.2.3,
it was assumed that the critical Category E′ welded cross-frame details were all governed by the
finite-life calculations and Fatigue II limit state, which is a reasonable assumption for Category E′
details and practical traffic conditions. Given that finite fatigue life is inherently a function of
both stress magnitude and frequency of load occurrence, the anticipated number of stress cycle
counts over the service life of these “fictional” bridges must be considered in some capacity.
The 4,104 bridges evaluated in this study, however, are not representative of any particular
location or traffic conditions. As such, the research team elected to investigate different repre-
sentative traffic conditions to bound the problem. In other words, these “fictional” bridges were
effectively constructed along different highway corridors in Texas, both in rural areas with
low traffic volumes and congested urban areas with high traffic volumes. Rather than utilize
measured WIM data (Section 2.3.3) at this stage, the research team utilized ADT maps readily
available on the TxDOT website (2020), as well as simplifying assumptions recommended
by AASHTO Article C3.6.1.4.2. A state highway system in rural Llano, Texas, and a heavily
trafficked corridor in Houston, Texas, were selected as the extreme conditions, but the general
process would be identical for other highway systems in other states.
Although not outlined in the report herein, AASHTO LRFD guidance was followed to obtain
an estimate of realistic single lane, average daily truck traffic (ADTTSL) values for the purposes
of computing the Fatigue II resistance stress ranges for the two extreme traffic conditions
considered. A detailed overview of these calculations is provided in Appendix F.
As noted previously in this section, the design stress ranges and resistances (as computed
by the procedures outlined above) are used throughout Chapter 3 to establish generalized
observations about load-induced cross-frame behavior in composite systems. More specifically,
these analytical data are processed and investigated in several different ways, as listed below.

Note that many of these concepts were previously introduced in Section 1.2 as the major
questions to be answered in NCHRP Project 12-113.
• Maximum stress ranges are compared for the various bridge geometries, such that the
impact of each parameter can be evaluated with respect to cross-frame behavior.
• Maximum stress ranges are compared for the various cross-frame members, such that
generalized observations about which members typically govern fatigue design can be made.
• The governing lane positions corresponding to those maximum stress ranges are compiled
for each representative bridge, such that the AASHTO fatigue loading model can be assessed in
terms of truck positioning.
• Factored stress ranges (based on current AASHTO LRFD fatigue loading criteria) are compared
to factored resistances, such that the overall efficiency of cross-frame design can be assessed.
• Factored stress ranges, determined by analytical methods, are compared to effective and
maximum stress ranges obtained from WIM records and the field experiments, such that the
accuracy and appropriateness of the current Fatigue I and II load criteria for application to
cross-frames can be assessed.
2.3.3 WIM Records

While Section 2.3.2 discussed the use of the AASHTO design load (i.e., the fatigue design
truck) and its design implications related to cross-frame force effects, this section discusses
the application of measured WIM data obtained from sites throughout the United States to a
subset of the 4,104-model data set. These analyses provided the opportunity to understand the
behavior of critical cross-frames for a representative set of bridge systems subjected to a variety
of realistic traffic conditions. This study of WIM data is intended to provide an indication on
the appropriateness of the current AASHTO LRFD Fatigue I and II load factors for the design of
cross-frames, as these load factors were developed primarily for girder response to truck traffic
per Modjeski and Masters (2015). In addition, this study investigates whether it is necessary to
consider multiple presence effects in the fatigue evaluation of cross-frames.
The research team obtained high-resolution WIM records from the Federal Highway
Administration (FHWA) for 16 specific pavement study (SPS) sites across the United States.
The records, collected in 2014, generally include a full year’s worth of measurements. The record
timestamps are reported with 0.01-second measurement resolution. In total, the unfiltered
records include approximately 46 million vehicle records from 16 sites over 15 states. Since
some sites have multiple lanes, the records include 23 one-lane records.
Consistent with SHRP 2 Project R19B (Modjeski and Masters 2015), the research team
applied a set of filtering techniques to this data set in an attempt to eliminate questionable
records (i.e., unrealistic geometry or erroneous data), apparent permit vehicles or illegally
loaded vehicles, and lightweight vehicles [i.e., gross vehicle weights (GVWs) less than 20 kips].
Many of these filters were based on NCHRP Research Report 683 (Sivakumar, Ghosn and
Moses 2011). Appendix F summarizes these filters. The lightweight vehicle entries were elimi-
nated, since previous research has indicated that these vehicles have a negligible effect on the
accumulated fatigue damage in a member or detail (Connor and Fisher 2006). White (2020)
conducted multiple sensitivity studies using the same records studied by the research team,
in which the lightweight vehicles were retained in order to study the impact on force effects
in cross-frames. The results of these studies indicated the inclusion of lightweight vehicle
records contributed negligibly to the accumulation of fatigue damage.
Table 2-3 provides a summary of WIM records prior to any filtering, as well as the total number
of vehicle records and ADT counts for each SPS site before and after removing lightweight

Table 2-3. SPS sites from which WIM data was obtained by FHWA.
Records Including Light Records After Removing Light

Initial Number
Vehicles Vehicles
of Records
State
(Before
Total Number Total Number of
Filtering) Lane ADT Lane ADTT
of Records Truck Records
AR 3,529,952 3,414,934 9,356 1,704,481 4,670

AZ 2,711,532 2,626,954 7,704 1,227,567 3,600
CA 4,873,640 4,779,602 13,167 1,380,075 3,802
CO 1,675,744 1,645,722 4,509 352,198 965
IL 2,807,183 2,707,469 7,584 798,935 2,238
IN (Lane 1) 1,886,428 1,865,543 5,111 370,241 1,014
IN (Lane 2) 500,621 471,291 1,302 21,340 59
IN (Lane 3) 1,843,395 1,802,053 4,937 360,458 988
IN (Lane 4) 548,382 524,003 1,460 20,158 56
KS 2,312,975 2,262,526 6,199 436,913 1,197
LA 1,734,519 1,722,311 4,719 76,547 210
MD 3,040,831 3,005,933 8,564 108,881 310
MN 864,803 857,522 2,349 52,757 145
NM1 1,018,250 1,005,887 2,786 147,077 407
NM2 1,675,090 1,609,947 4,485 892,295 2,486
PA 2,292,235 2,251,316 8,187 873,903 3,178
TN (Lane 1) 2,792,715 2,120,750 7,913 550,858 2,055
TN (Lane 2) 1,945,926 1,842,295 6,874 118,000 440
TN (Lane 3) 2,042,591 1,942,114 7,247 191,330 714
TN (Lane 4) 2,757,569 2,690,197 10,038 1,182,136 4,411
VA (Lane 1) 1,462,016 1,445,614 3,961 224,928 616
VA (Lane 2) 438,126 430,438 1,179 19,300 53
WI 1,468,798 1,358,660 5,435 120,079 480
Total 46,223,321 44,383,081 11,230,457
vehicles. The records used for the fatigue study contain approximately 11 million truck measure-
ments after all appropriate filters are applied.
Figure 2-7 shows the cumulative distribution functions (CDFs) of GVWs captured by the
WIM sensors for all SPS sites. The GVWs are shown on a normal probability plot, in which
the horizontal axis is GVW (in kips), and the vertical axis is the standard normal variable
(i.e., axis values are “Z-values” indicating the number of standard deviations the GVW value
is from the mean of the distribution). A normal probability plot can be used to determine how
well the data represents a normal distribution; nonlinear data sets indicate departures from
normality. As clearly shown, the GVW populations are not normally distributed. This likely
is the result of natural groupings of different vehicle types and payload. The mean GVWs for
the 11 million truck records range from 40 to 62 kips, with a maximum GVW of 220 kips. As
documented in Appendix F, the shape of the CDFs appears to be generally consistent with
the CDFs of WIM data used in the SHRP 2 R19B project (Modjeski and Masters 2015), which
were recreated by the research team with the same filters described above. Thus, the 2014
data obtained from FHWA that were used on this project were deemed acceptable given their
good agreement with the SHRP 2 R19B data.

2
Standard Normal Variable
-1
-2
-3
-4
-5
0 50 100 150 200 250
GVW (kips)
Figure 2-7. CDF of GVWs from FHWA 2014 data (excluding light vehicles).
2.3.3.1 Application of WIM Records

In order to estimate real live load force effects on cross-frame members, the research team
applied the filtered WIM traffic streams to a subset of the analytical testing matrix discussed in
Section 2.3.1.2. As the computational effort is significant (discussed in the next subsection),
approximately 20 models were selected from the 4,104-model data set. While the models were
somewhat arbitrarily selected, bridge models that would likely represent extreme cross-frame
force effects based on preliminary studies were chosen. Note that these same representative
bridges were subsequently evaluated in the parametric studies related to the R-Factor Study
outlined in Section 2.4 and the Commercial Design Software Study outlined in Section 2.5.
2.3.3.2 Processing of WIM Data

Using MATLAB, a script was developed to read, format, and filter the entire WIM record
database. Since the axle tracks (i.e., the transverse distance between the centerline of two wheels
on the same axle) are not recorded in the WIM records, the research team set all axle tracks
to 6 feet and assumed that the weight of each axle is evenly distributed between the driver and
passenger side wheels.
The research team created scripts to perform the following basic load configuration routines:
• Load Configuration 1 – Single Traffic Stream: This routine steps a stream of user-defined WIM
traffic on a bridge deck along an influence surface at 1-foot longitudinal increments in any
defined transverse position (also 1-foot increments). The script uses a cluster analysis to
include the effects of groups of vehicles in the same traffic stream, provided that any wheel
of the following truck is on the bridge during the time window in which the leading vehicle
is still on the bridge. Time windows are calculated based on the respective vehicle speeds.

• Load Configuration 2 – Two Traffic Streams: Similar to Load Configuration 1, a WIM traffic
stream is stepped along a bridge deck in a defined transverse position. The script uses a cluster
analysis to include the effects of groups of vehicles in any adjacent, user-defined transverse
position, provided any tire on the following truck is on the bridge during the time window
in which the leading vehicle is still on the bridge. Time windows are calculated based on the
respective vehicle speeds.
• Load Configuration 3 – Realistic Meandering Traffic Stream: This routine systemically steps
a stream of WIM traffic along a defined influence surface at 1-foot longitudinal increments
in any defined 12-foot-wide lane position. The routine randomly selects a transverse position
(within the 12-foot-wide lane) for each vehicle record based on an assumed distribution,
such that effects of lane meandering are considered due to cross-frames being highly sensitive
to transverse position. This script uses a cluster analysis to include the effects of groups of
vehicles, provided any of a following vehicle tires are on the bridge during the time window
that a leading vehicle is still on the bridge. Time windows are calculated based on the respec-
tive vehicle speeds.
Recalling that the influence-surface output from Abaqus provides results that are relatively
sparsely gridded (Section 2.3.1.4), the scripts use bi-linear interpolation methods to re-mesh
the influence surfaces to a 1-foot by 1-foot grid. This procedure is similar to that described
in Section 2.3.1 for the single AASHTO fatigue truck.
The output of each load configuration routine above includes the following:
• A sample load event history (see Figure 2-8);
• The total number of stress cycles (using rainflow-counting techniques discussed in Appendix F);
• The average number of cycles per passage;
• The maximum stress and stress cycle recorded;
• The equivalent stress range using the Palmgren-Miner damage accumulation model (i.e.,
stress range corresponding to the Fatigue II limit state); and
• The lowest stress range of the top 0.01% of all stress ranges (i.e., the 99.99th percentile criteria
corresponding to the Fatigue I limit state).
Past research has indicated that eliminating smaller stress cycles of a variable-amplitude
loading source has negligible effect on the damage accumulated in a fatigue detail. Connor
and Fisher (2006) showed that stress cycle magnitudes less than 25% of the constant ampli-
tude fatigue limit (CAFL) generally have little impact on the long-term fatigue performance,
2.0
Cross-Frame Axial Stress (ksi)
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0
2000 2200 2400 2600 2800 3000 3200
Load Step (1-ft Increment)
Figure 2-8. Sample load event history showing three vehicles
back to back.

which is outlined in Appendix E. To study the effects of this lower stress range truncation
on cross-frames, the research team compiled the results of each loading iteration (as listed
above) for two different conditions: (i) including the effects of stress magnitudes less than
25% of the CAFL and (ii) filtering out those effects. For the purposes of this study, a detail
Category E′ is assumed based on the results of McDonald and Frank (2009) and AASHTO
LRFD Table 6.6.1.2.3-1 (Condition 7.2). The corresponding CAFL value for this detail is 2.6 ksi;
thus, all stress cycles less than 0.65 ksi are truncated when the filter is applied. For the remainder
of this report, the data discussed refers to the unfiltered data (i.e., no stress ranges are truncated).
A more detailed discussion of the effects of stress truncation is provided in Appendix F.
The routines discussed in the previous section systematically and sequentially apply a defined
traffic stream in a specific load configuration over the influence surface corresponding to one
cross-frame member in one of the 20 representative bridges. Considering the computational
time for each iteration (depending on the total number of vehicle entries and bridge length) the
number of potential iterations becomes unmanageable. For this reason, the studies presented
in this report were obtained by positioning the WIM traffic stream in a realistic drive lane,
rather than a worst-case drive lane.
The worst lane position is often when the truck is positioned at the edge of the deck (e.g., the
right tire of a truck is positioned along the deck edge where a traffic barrier would be placed).
As a worst-case lane position would lead to unnecessary conservatism, the research team opted
for defining a realistic lane position based on the actual bridge width and recommendations
from AASHTO Article 3.6.1.1.1. As illustrated in Figure 2-9, this realistic truck position does
not produce the largest tensile stress range for the selected cross-frame; however, this position
would be consistent with realistic loading and lane striping on the bridge (and thus consistent
with the intent of the fatigue limit state).
With that in mind, the research team deemed Loading Configuration 3 introduced above
to be more pertinent with respect to fatigue design. As such, the focus of the study presented
herein is on realistic traffic streams as opposed to considering all possible lane positions (i.e.,
overhang loads).
Figure 2-9. Illustration of realistic truck position in drive lane.

For the actual location of each 6-foot vehicle track width (i.e., transverse distance between
left and right wheel lines) within the realistic lane, the research team assumed a distribution for
which the vehicle is located in the center of the lane 55% of the time; 30% of the time, the vehicle
is located plus or minus 1 foot of the lane centerline; 10% of the time, the vehicle is located plus
or minus 2 feet of the lane centerline; and 5% of the time, the vehicle is riding along one
of the lane edges. This is illustrated in Figure 2-10 for a sample 30-foot wide bridge, although
the procedure is similar for different bridge widths and lane configurations. The intent of this
assumed distribution, although not based on past precedents or measurements, is to consider
the inherent variability in driving ability and consistency.
2.3.3.3 WIM Drive Lane Truck Traffic Positioned in Realistic Drive Lanes
The traffic streams discussed herein exclude vehicles with GVW less than 20 kips and utilized
a drive lane defined by AASHTO Article 3.6.1.1.1, along with an assumed vehicle position
distribution discussed in Section 2.3.3.2.
For Fatigue I, the maximum stress ranges produced by the WIM traffic streams were normal-
ized to the maximum stress range produced by applying the unfactored AASHTO design load
(i.e., HS-20) to the same bridge in all possible transverse positions within the clear distance of
the barriers. An impact factor was not included since the WIM stations attempt to correct
for dynamic effects and relate measured drive-by weights to static weights. Note that it may
be prudent to include a portion of the typical 0.15 impact factor (AASHTO Article 3.6.2.1)
in subsequent studies, since the effects of bridge dynamics are not accounted for explicitly in
this FEA.
The governing Fatigue I maximum stress ranges calculated for each individual WIM site
and bridge combination were determined based on a 99.99th percentile criteria. This stress
range corresponds to the lowest magnitude of the top 0.01% of all stress ranges recorded (for
the governing cross-frame for each bridge given the defined lane position). In other words,
one critical cross-frame produced WIM stress ranges that exceeded the remainder of the cross-
frames in the bridge and thus served as the governing case.
Similarly, the governing Fatigue II stress ranges calculated for each WIM site and bridge
combination were determined based on the maximum effective stress range. Recall that
the Fatigue II limit state is intended to represent the average effect of the traffic spectrum.
These effects are generally characterized by the effective stress range (or equivalent stress range)
of the variable-amplitude response. Appendix F discusses how the effective stress range is
Figure 2-10. Illustration showing distribution of

vehicle transverse location within 12-foot design lane
for a 30-foot wide bridge.

obtained from a variable-amplitude spectrum via rainflow-counting techniques and the use of
Palmgren-Miner’s rule. In order to compare the WIM force effects to the unfactored AASHTO
design truck force effects, it is more appropriate to compare the accumulated fatigue damage,
rather than just the effective stress ranges. The accumulated fatigue damage metric inherently
considers both the variable stress range magnitudes and the number of cycles. Thus, the total
damage accumulated by the various WIM streams on the critical, governing cross-frame
members is compared directly to the damage caused by the AASHTO fatigue truck.
2.3.3.4 Reliability Study Using Proposed Load Factors

A reliability study was performed to investigate the implications of using the load statistics
developed in Section 2.3.3.3 for fatigue of cross-frames in steel I-girder bridges in the context
of a reliability-based design. Using the statistical characterizations of the total load effect, it
is possible to compare this to the statistical characterization of fatigue resistance provided in
the literature (Modjeski and Masters 2015). Stochastic models can be implemented that utilize
these characterizations of load and resistance to provide an indication of reliability, given a
specific choice of design parameters.
The reliability study is performed via Monte Carlo simulation, in which values of load and
resistance are obtained through a randomly determined process using distribution parameters
for load and resistance. For this reliability study, the research team adopted the resistance
models developed in the SHRP 2 R19B report (Modjeski and Masters 2015). Appendix F
provides the step-by-step procedure used in applying the Monte Carlo simulation. Results
of the Monte Carlo simulations are subsequently summarized in Section 3.2.2.3.
2.3.3.5 Multiple Presence Study

The WIM data set used in this study and outlined in Section 2.3.3 includes multi-lane records
for three sites. As two of the sites (Indiana US-31 and Tennessee IH-40) include two lanes of
traffic in two directions, the multiple presence data set includes a total of five 2-lane records.
In order to determine if a passing truck is in the same or an adjacent lane as another truck, it is
necessary that the time stamps have high resolution. For example, many available WIM records
have a time resolution of 1 second. At 70 miles per hour, a time difference of plus or minus a
half-second is equivalent to a range in truck position of approximately plus or minus 50 feet,
making assessment of multiple presence for cross-frame fatigue highly unreliable. Higher reso-
lution data is essential to evaluate the frequency of occurrence of multiple presence. At 70 miles
per hour, the 0.01 second resolution data that the research team obtained from the FHWA
generally provides a resolution of plus or minus 0.5 feet.
With the availability of this higher-resolution, multi-lane data, the team performed a study
on the WIM records to better understand the statistical parameters surrounding multiple
presence. This was performed using a cluster analysis to consider if a bridge may be loaded
with other simultaneous truck traffic during a primary drive lane load event (i.e., passage of
one or more axles of a vehicle). The cluster analysis is performed based on the time stamps of
the individual truck events, the lengths of the individual trucks, and the speed of the individual
trucks. Since bridge lengths vary, the following study incorporates multiple presence load events
that occur within a plus or minus 1,000-foot window of the primary drive lane event.
The research team developed a script to determine how many times a second truck is in a
lane adjacent to the primary drive lane truck (i.e., a second truck is passing or is being passed
by the truck in the drive lane, anywhere along an arbitrary length of 1,000 feet). This scenario
is illustrated schematically in Figure 2-11. Clear distances are measured from the rear axle of
the drive lane truck to the front axle of the passing lane truck. Positive values indicate front
axles of the passing truck are “behind” the rear axles of the drive-lane truck, and negative values

Passing Truck
Drive Lane Truck Negative Clear

Distance
Passing Truck
Positive Clear Drive Lane Truck

Distance
Figure 2-11. Illustration of an adjacent lanes (truck-passing-truck)

scenario; negative and positive clear distances indicate the front axle of
the passing truck is ahead or behind the rear axle of the drive lane truck,
respectively.
indicate the front axles of the passing truck are “ahead of ” the rear axles of the drive-lane truck.
This provides a smooth, continuous function of clear distances, with increasing clear distances
(positive or negative) indicating a larger separation between vehicles.
Note that a clear distance of zero corresponds to a staggered configuration, which was
deemed critical to cross-frame force effects in the 7th Edition AASHTO LRFD Specifications.
This provision was removed in the 2016 Interims to the 7th Edition Specifications, citing too
infrequent occurrences. This study herein evaluates the frequency of occurrence in the context
of cross-frame response. Results of the multiple presence study are subsequently summarized
in Section 3.2.3.
2.4 R-Factor Study (3D Analysis)

The R-Factor Study, as noted previously, addresses Objective (c) of this project (Section 1.2):
to provide quantitatively based guidance on the influence of end connections on cross-frame
member stiffness. With that in mind, the primary focus of this study is the 3D modeling
techniques commonly employed by commercial design software programs. While the term
“3D models” are often synonymous with highly accurate representations of the framing system,
the level of detail in modeling the cross-frames can have a dramatic impact on the accuracy of
the brace forces estimated. Commercial 3D programs generally simplify cross-frame members
and their connections as pin-ended truss elements, thereby misrepresenting the stiffness of
the panel and potentially producing unreliable design force estimates.
As noted in Section 1.1.3, significant research has been conducted over the past decade to
approximate the softening effects of the eccentric load path through unsymmetrical cross-
frame members (Wang 2013; Battistini et al. 2016). Although the previous work developed
equations for evaluating the reduction in stiffness, the recommendations incorporated into the
design provisions consist of a fixed stiffness modification factor of 0.65 (AASHTO LRFD Article
C4.6.3.3.4). However, these past studies were primarily limited to the response of cross-frames
in noncomposite systems during construction. As such, the stiffness response of cross-frames
in composite in-service bridges has not been properly investigated. This section provides an
overview on the computational studies performed to evaluate the limitations of the traditional

truss-element modeling approach for cross-frames, appropriate stiffness modifications for

composite conditions, and a proposed alternative approach to model cross-frames in refined
3D analyses (i.e., an eccentric-beam model).
Section 2.4.1 outlines the general load-induced behavior of cross-frames in composite sys-
tems to provide context to the analysis procedures and results presented herein. Section 2.4.2
highlights key aspects of the stiffness modification approach, and Section 2.4.3 outlines the
proposed eccentric-beam modeling approach. Lastly, Section 2.4.4 summarizes the parametric
studies performed to evaluate each of these methodologies.
2.4.1 General Load-Induced Cross-Frame Behavior

Prior to examining the various 3D modeling techniques, it is important to understand
how cross-frames deform under applied live loads in composite systems. By comparing “real”
deformations with the deformations assumed in common 3D analyses, the remaining discus-
sions of this section can be contextualized.
Intuitively, the relative movement of the corners of a cross-frame dictates the axial-force
demands in each member. During this study, the observation was made that any load-induced
deformed shape in a composite system is a combination of the following responses between
adjacent girders framing into a common cross-frame: (i) equal rotation, (ii) differential vertical
displacement, and (iii) differential rotation. These deformations are presented schematically
in Figure 2-12. The cross-frame also undergoes rigid-body motion as the entire superstructure
displaces under applied load. However, rigid-body motion does not induce strains and stresses
on cross-frame members and is therefore not relevant to the discussion herein. The previous
work reported in Battistini et al. (2016) and Wang (2013) focused only on the case of equal
rotation, which resulted in the fixed reduction factor of 0.65 recommended in AASHTO LRFD
Article C4.6.3.3.4.
Even in Figure 2-12, the deformation of the cross-frame members is idealized (i.e., only
axial deformations are represented). In reality, the entire panel (members, gusset plates, and
connection plates) undergoes some degree of in-plane and out-of-plane rotation under an
applied load. Consequently, cross-frame members experience biaxial bending stresses in addi-
tion to axial stresses. A schematic of this behavior is presented in Figure 2-13 for a scenario
in which the girders sustain purely differential vertical displacements. A cross-sectional view
and a plan view are provided to examine the in-plane and out-of-plane behavior, respectively.
The column depicting the shell-element models indicates that FEA models comprised of
shell elements accurately capture these rotational effects, compared to the column depicting
truss-element models.
Differential Vertical
Equal Rotation Differential Rotation
Displacement
θy θx2
θx
θx θx1
θy
Figure 2-12. Types of cross-frame deformation caused by relative

movement of adjacent girders.

View Shell-Element Models Truss-Element Models
In-
Plane
Out-of-
Plane
Figure 2-13. Schematic showing the in-plane and out-of-plane

flexural deformations commonly observed in shell-element models
compared to simplified truss-element models.
Figure 2-13 divides the deformations into in-plane and out-of-plane deformations. The
in-plane rotations depicted in Figure 2-13 are attributed to the in-plane rigidity typically provided
by the connection and gusset plates. In order to accommodate the girder displacements and
compatibility of the system, the cross-frame members have to flex given the rotational restraint
provided by the end connections. Given that the in-plane rotational stiffness of the connection
plate and welds is generally large (i.e., full-depth web stiffener welded along three sides),
this behavior is likely observed in all practical steel I-girder systems in service. This behavior,
however, is typically ignored in the truss-element modeling approach.
Out-of-plane rotations, on the other hand, are attributed to the eccentric nature of the connec-
tions. As internal force is transmitted from the girder webs into the cross-frame members, the
load must pass several “eccentric jumps.” These “jumps” of potentially several inches in magnitude
can have a considerable impact on the stiffness response of cross-frame members and systems.
The stiffness modification factors (R-factors), detailed in the next subsection, were primarily
developed to account for this modeling oversight in truss-element models.
2.4.2 Stiffness Modification Approach

As indicated above, most 3D commercial design software programs implement a truss-
element approach for modeling cross-frames. Rather than explicitly modeling each component
and the load eccentricity of the cross-frame with shell elements, commercial design software
programs typically simplify them as pin-ended truss elements capable of only resisting axial
forces. It should be noted that even 3D models utilized in many past research investigations
have idealized the cross-frames using truss elements. In this simplified approach, the eccentric
load path and the stiffness contributions of the gusset and connection plates are neglected, as
are the biaxial bending stresses induced. In general terms, there are three inherent shortcomings
with this modeling approach when compared to realistic cross-frame systems or a shell-element
modeling approach for cross-frames (two of which were illustrated in Figure 2-13). These
shortcomings are summarized below:
• The truss-element modeling approach neglects the out-of-plane bending effects caused by the
eccentric end connections. At the most fundamental level, the eccentric load path in an
unsymmetrical cross-frame member is introduced by the connections. The stiffness modi
fication factors established by Battistini et al. (2016) were directly developed to account

for the softening effect of eccentric end connections (i.e., the R-factor is often assigned to
reduce the cross-sectional area of the truss element or the elastic modulus of the material).
As such, this aspect is a well-researched behavior.
• The truss-element modeling approach neglects the in-plane bending effects caused by the connec-
tion rigidity. Unlike the out-of-plane effects, the stiffness modification factors formulated
by Battistini et al. (2016) did not explicitly address these effects. Instead, in-plane rotational
behavior is implicitly considered in those previous studies.
• The truss-element modeling approach neglects the axial stiffness of the connection and gusset
plates in favor of the cross-frame member axial rigidity. In other words, truss-element models
replace the stiffer connection and gusset plates with an additional length of an often, more
flexible cross-frame member. Again, this behavior was not explicitly considered in the
development of the R-factors but was implicitly considered in the corresponding laboratory
and analytical studies.
To expand on the previous work, two generalized solutions were explored in NCHRP
Project 12-113. First, stiffness modification factors similar to those derived by Battistini et al.
(2016) were analytically developed in the context of composite bridge systems. To accomplish
this, a series of 3D analyses in Abaqus and approximate hand calculations were conducted at
three different model-scale conditions: (i) member-level, (ii) panel-level, and (iii) system-level
studies. The second generalized solution is a proposed alternative to the truss-element modeling
approach that makes use of eccentrically loaded beam elements for the cross-frame members.
These procedures are briefly outlined herein.
2.4.3 Proposed Eccentric-Beam Approach

As noted previously, the truss-element modeling approach has three major limitations with
regard to representing the in-plane flexural, out-of-plane flexural, and axial rigidity of cross-
frame systems. The stiffness modification (R-factor) approach corrects for those limitations by
approximately adjusting the axial rigidity of a pin-ended truss element such that the overall
cross-frame stiffness in the analysis model reasonably matches its actual stiffness. As with many
specifications and design guides around the world though, it is often beneficial to provide
designers with multiple solutions to a problem with varying levels of refinement.
As an alternative, the research team sought an approach more refined than using pin-
ended truss elements, yet simpler to employ than the shell-element approach. The proposed
eccentric-beam modeling approach addresses the major limitations of the truss-element
approach, similar to assigning R-factors but in a more explicit and less circuitous manner.
The proposed methodology represents the cross-frame members (and connection and gusset
plates) as beam elements with flexural degrees of freedom. Note that the eccentric-beam
approach does not necessarily correspond to one specific technique but rather encompasses a
wide range of modeling assumptions with varying levels of accuracy and modeling simplicity,
as demonstrated below.
The eccentric-beam approach is shown schematically in Figure 2-14 along with a representa-
tion of the shell-element and truss-element approaches for cross-frames in 3D models. Unlike
the truss-element approach, in which the cross-frame stiffness is represented as only the axial
rigidity of the cross-frame member modified by an R-factor (i.e., RAa), the eccentric-beam model
is represented by a larger set of parameters, including but not limited to –y , the out-of-plane
eccentricity, and Aa, the cross-sectional area of the cross-frame member (not modified by R).
The biggest challenge with respect to the eccentric-beam approach is deciding how/if to
handle the axial and flexural stiffness of the connection plate, the gusset plate, and the over-
lapped portions. As such, three different variations of the approach were explored, ranging

Pin-Ended Truss-Element
Shell-Element Eccentric-Beam
with R-Factor
y–
Aa
RAa
Figure 2-14. Various cross-frame modeling approaches

considered in system-level studies.
from more refined to relatively simple. Each variation (i.e., the refined, simplified, and angle-
only models) is presented schematically in Figure 2-15. This figure schematically depicts
these different modeling methods for a sample cross-frame panel. Note that the “eccentric
jumps” only apply to the out-of-plane direction, as the behavior is assumed concentric in
the in-plane axis.
To illustrate some of the assumptions used in the development of these models, an abbre-
viated set of notes is provided below. This discussion specifically highlights key aspects of the
“refined” model, as defined by Figure 2-15, but are also directly applicable to the “simplified”
and “angle-only” approaches, where appropriate. For a more detailed overview, as well as
sample calculations demonstrating the proposed methodologies, refer to Appendix F.
• In the “refined” model, the cross-frame member, gusset plate, connection plate, and
individual overlapped sections are treated independently when developing the eccentric-
beam model.
96″
Connection plate
Overlap 1
Gusset plate
Overlap 2
L4x4x3/8
60″ Girder web shell (typ.)
Refined Model
Connection plate
L4x4x3/8
8″
Simplified Model
9/16″ PL.
4″
1″ L4x4x3/8
L4x4x3/8
16¼″ ½″ PL.
Pinned or fixed
L4x4x3/8
Angle-Only Model
7¼″
Note: Out-of-plane view shown; all
“offset” links are rigid elements
Figure 2-15. Sample cross-frame panel and the corresponding

eccentric-beam models.

• The out-of-plane “jumps” in the neutral axis, or the eccentricities, are represented by rigid
offsets, which imply that the welded or bolted connections between the plates and angles
perfectly constrain these elements together (i.e., no slip is assumed in the bolted connections).
• The lengths of the rigid elements are based on the distances between the neutral axes of the
connected components.
• The length of each individual beam component is assigned assuming there are no shear lag
effects. In other words, the length of the angle member starts precisely at the termination of
the gusset plate and ends at the gusset plate edge on the opposite side of the angle.
• Section properties (i.e., A, out-of-plane I, in-plane I, and length L) of the connection and
gusset plates are based on a Whitmore approach. Section properties of the equivalent
overlapped portions are taken from an assumed composite section. A detailed calculation
demonstrating this process is provided in Appendix F.
• Unlike the pin-ended truss models, rotations are not released at both ends of the eccentric-
beam model. The distortional stiffness of the girder web is assumed substantial enough to
warrant in-plane and out-of-plane rotational restraint of the cross-frame system.
• In X-type cross-frames, the diagonal members are connected on opposite faces of the cor
responding gusset plates to avoid interference at the crossover point. These different
eccentric effects were considered accordingly.
• Although not shown here, a similar procedure was also used to develop equivalent beam
properties for K-type cross-frames.
This proposed 3D modeling technique was evaluated for a variety of bridge geometries,
as outlined in the next subsection.
2.4.4 Parametric Study Overview

To evaluate the various simplified 3D cross-frame modeling techniques, three different
model-scale conditions were identified and studied including: (i) member-level, (ii) panel-level,
and (iii) system-level studies. The work of each successive study built on the previous one.
Member-level studies computationally investigated the fundamental behavior of axial-loaded
sections with eccentric end connections, similar to the experimental approach by McDonald
and Frank (2009). Panel-level studies analyzed isolated cross-frame panels consisting of
various geometries, as well as various gusset and connection plates. The objective of the system-
level study was to demonstrate how the behavior of cross-frames (i.e., load-induced force
demands) is affected when serving as a small part of a larger composite system. A parametric
study consisting of full 3D bridge models was conducted to assess the accuracy of the stiffness
modification factors, as well as the eccentric-beam approach.
To elaborate on these studies, two subsections are provided herein. An overview of the
panel-level computational studies is provided in Section 2.4.4.1, while the system-level studies
are discussed in Section 2.4.4.2. Note that the member-level studies are not explicitly addressed
in this report, but the pertinent background and results are provided in Appendix F.
2.4.4.1 Panel-Level Studies (Noncomposite)

The panel-level studies, as noted in the preceding section, examined the elastic stiffness
response of individual cross-frame panels. First-order finite element models were developed
to duplicate the laboratory experiments and analytical studies previously conducted by Wang
(2013). However, the research team expanded on those past investigations by subjecting
the panels to a variety of deformation patterns, as illustrated in Figure 2-12. Note that the
previous studies, which focused on stability bracing applications, were limited to the equal
rotation deformation pattern to simulate the buckling and torsional response of straight and
horizontally curved noncomposite bridge girders during construction.

In general, the behavior of a shell-element model is compared to the equivalent truss-

element model. The ratio of the corresponding stiffness of a shell-element model to that of a
truss-element model represents the R-factor. The shell-element model explicitly considers the
reduction in stiffness due to eccentric end connections and the effects of additional in-plane
rotational restraint, while the truss-element model considers only axial stiffness. The elastic
stiffness of each model iteration was taken as the torsional stiffness of the entire cross-frame
panel, or the rotational displacements due to an applied moment or twist on the entire panel.
Prior to parametrically studying the effects of different cross-frame geometries, member
sizes, and connection details, the research team conducted preliminary studies to verify the
modeling approach and to achieve good agreement with measured results from past experi-
mental tests. For instance, considerable effort was spent on connection restraints (i.e., how
welded connections were represented computationally), which were ultimately found to have
only a small influence on the cross-frame response.
In the parametric studies, various cross-frame parameters were systematically evaluated
through a series of FEAs. The variables considered in the study included girder spacing,
cross-frame height, gusset/connection plate thickness, and angle sizes. The range of values
considered for each parameter represented practical values as well as additional extreme condi-
tions to bound the solutions. Note that the previous analytical studies by Wang (2013) and
Battistini et al. (2013) did not explicitly consider the effects of connection plate thickness.
A thicker connection plate implies two contrasting behaviors: a larger eccentricity but also
a larger out-of-plane rotational stiffness. Therefore, this study examined how the stiffness
modification factor is influenced by these two factors.
In total, approximately 2,000 unique cross-frame geometries were developed as part of this
parametric study. The general-purpose 3D FEA program Abaqus was used for the parametric
studies. Refer to Appendix F for a detailed overview of the parameters considered.
The researchers found that a key factor in the behavior of the panel-level analytical studies
was related to the orientation of the cross-frame member, specifically the diagonals, relative
to the gusset and connection plates. For X-type cross-frames, the diagonal members must be
connected to opposite faces of the gusset plates to avoid interference at the crossover point.
The position of the struts can also vary and depends on the standard DOT details and fabri
cator preference. Based on preliminary analyses, it was discovered that the orientation of the
member connection had substantial impact on the stiffness response of the panel, depending
on the deformation pattern and the distribution of forces. This is depicted schematically in
Figure 2-16, which presents a detailed close-up of the top strut and diagonal connection for
different loading conditions.
In the right side loading condition, the top strut and the grey diagonal member are directly
engaged and provide the primary resistance to the applied force. These two members are
connected to the same face of the gusset plate. In this case, the corresponding eccentricity of this
connection is additive with respect to the gusset-to-connection plate connection. Conversely
in the left side loading condition, the top strut and the yellow diagonal member are directly
engaged and provide resistance to the applied force. The diagonal in this scenario is fastened
to the opposite face of the gusset plate as the top strut. Therefore, the individual member
eccentricities tend to counterbalance each other. This, in turn, often results in a stiffer response
by reducing the net eccentricity. This particular behavior is quantified in Section 3.3.
2.4.4.2 System-Level Studies (Composite)

Whereas the panel-level studies evaluated the global stiffness of an isolated cross-frame, the
system-level study assessed the accuracy of the simplified modeling approaches for 3D composite

Figure 2-16. Assumed orientation of cross-frame members for differential-

rotation loading.
bridge structures with various gusset plate thicknesses. More specifically, the sensitivity of
the load-induced cross-frame response due to the assigned R-factor was examined.
To make those assessments, a variety of analytical studies were conducted as part of the
system-level study. A parametric study (consisting of only straight and normal bridges) was
performed to emphasize the influence of connection and gusset plate dimensions on this
behavior. Additionally, a second study was conducted to further investigate the effects of
support skew and horizontal curvature on the R-factor and eccentric-beam approaches.
For each individual study, the research team developed and compared multiple versions of
the same 3D bridge model. Each iteration investigated a different approach to modeling the
cross-frame elements, while the remainder of the 3D model, as outlined in Section 2.3.1, remained
unchanged. In general, the rigorous shell-element model served as the benchmark model to
which the rest of the iterations were compared. Truss-element models were also developed where
the assigned R-factor was varied. Lastly, the proposed eccentric-beam modeling approach was
also implemented for the various bridge models. Thus, for each individual study, the following
model iterations were considered:
• Shell-element model (representing the most accurate solution),
• Truss-element model with different stiffness modification factors {R = 0.5, 0.6, 0.7, 0.8,
0.9, 1.0}, and
• Eccentric-beam model.
For the shell-element and eccentric-beam modeling approaches, different values of the gusset
plate thickness were also considered as a variable. Note that the gusset plate thickness and the
corresponding effects on the load eccentricity were explicitly considered in the shell-element
model. In the eccentric-beam model, the eccentric offset dimension, as illustrated in Figure 2-15,
and the equivalent beam properties were adjusted accordingly. Gusset plates were not explicitly
represented in the truss-element models.
Ultimately, the effectiveness of the simplified 3D model (i.e., the truss-element model with
an assigned R-factor or the eccentric-beam model) is evaluated as a ratio between the predicted
axial force in select cross-frame members from the simplified models to the predicted force in
the same members from the shell-element model. This is described algebraically as Fsimplified/Fshell.

Rather than compare the stiffness of the models indirectly, it is more appropriate to evaluate the
ability of the simplified model to accurately obtain design forces, which is of most importance
to designers.
A ratio of unity represents perfect agreement between the shell-element model and the
simplified models. This implies that the approximated stiffness is identical to the “true”
stiffness of the panel. Values below unity indicate that either: (i) the assigned R-factor is
likely too low for the truss-element model or (ii) the equivalent section properties assigned for
the eccentric-beam model are too low (i.e., the cross-frame attracts less force). In other words,
the simplified models underpredict the cross-frame force when compared to the more accurate
shell-element model (i.e., unconservative estimates), and an increase in the modification factor
or equivalent beam properties is needed. The opposite is true for force ratios above unity.
In all cases, the applied load on these composite bridge systems represented fatigue design
load conditions. Thus, the cross-frame force effects compared between the various modeling
approaches were based on the same corresponding load cases (i.e., same truck configuration,
weight, and lane position). For more information on the parameters considered and the perti-
nent loading conditions, refer to Appendix F. The results of this computational study are sub
sequently summarized in Section 3.3.
2.5 Commercial Design Software Study (2D Analysis)

As discussed extensively throughout this report, analysis refinement generally improves
accuracy but at the cost of increased modeling complexity and computational effort. As such,
many designers and commercial design software programs elect to use simplified methods
to conduct refined analysis and obtain design forces. These simplified methods range from
high-fidelity 3D FEA solutions, which were introduced in Section 2.4, to 1D line-girder analyses.
One-dimensional line-girder analyses can be useful tools for girder analysis and design of
simple structures but generally provide no information related to cross-frame behavior.
With that in mind, outlining the methodology behind the 2D analysis methods prevalently
used in the bridge design industry is the primary focus of this section [i.e., Objective (d) of
NCHRP Project 12-113]. Two-dimensional analysis methods have been shown to produce
relatively accurate results for girder forces (White et al. 2012), but that performance is less
understood and quantified for cross-frame forces, particularly in composite bridge systems.
In general, the Commercial Design Software Study examined the limitations of these common
2D methods with regards to cross-frame force effects in composite bridge systems through a
series of analytical studies.
As noted in Section 2.1, the research team polled bridge owners and consultants across
the country on their preferred commercial design software programs. Based on these results,
two different programs were identified for further investigation—one 2D program and one
3D program. Since the specific analysis techniques are of more importance to the scope of
the project than the software brand itself, it was decided to conduct all studies related to this
study in Software A, a widely used 3D analysis program. Although specifically developed and
marketed as a 3D bridge design tool, Software A is also capable of developing essentially the
identical 2D models associated with Software B.
By using Software A to conduct all 2D and 3D commercial design software analyses, two major
benefits were realized. First, rather than being restricted to the inherent “black box” assumptions
adopted by many 2D-specific software packages, manually developing 2D models in a general-
use program provided more flexibility, especially in terms of implementing several modeling
improvement methods (Section 2.5.3). Secondly, this general approach shifted the focus away

from comparing specific commercial design software packages and instead highlighted the
modeling approach, which is of more general value to designers and software developers.
In total, three distinct types of analyses were performed in Software A as part of this study,
including 3D models (as a direct comparison with the 3D truss-element models produced in
Abaqus and documented in Section 2.4), 2D plate and eccentric-beam (PEB) models, and 2D
grid/grillage models. The following subsections outline key attributes related to these 2D and 3D
modeling procedures. Section 2.5.1 briefly introduces the major assumptions and features related
to PEB and grillage models, as well as explains the subtle differences between the 3D models
developed in Software A and Abaqus (Section 2.4). Section 2.5.2 then expounds on the equiva-
lent beam approach for modeling cross-frames that is inherent with any 2D analysis. Finally, Sec-
tion 2.5.3 offers techniques commonly used to improve the 2D equivalent beam approach, and
Section 2.5.4 outlines the parametric studies used to evaluate these analytical techniques. Note
that the general approach for these simplified methods is largely based on the guidelines docu-
mented in AASHTO G13.1 Guidelines for Steel Bridge Analysis (2019) and White et al. (2012).
2.5.1 General Modeling Assumptions

The inherent modeling assumptions associated with 3D, 2D PEB, and 2D grillage models
in Software A (and many other commercial programs) are briefly outlined in this section.
Individual subsections are provided to discuss each approach.
2.5.1.1 3D Models
Three-dimensional modeling of bridge systems in any commercial design software (e.g.,
Software A) can vary from package-to-package and engineer-to-engineer. The accuracy of the
model as it pertains to cross-frame forces can be sensitive to the assumptions made by the engi-
neer. With this in mind, the intent was to provide a modeling approach that is representative
of most 3D software packages and assumptions likely utilized by bridge engineers. However, it
is important to note that variability in the results should be anticipated for a different software
package or a different set of assumptions.
The 3D models in Software A were developed similarly to the 3D Abaqus model (Section 2.3.1)
with a few notable exceptions. First, girder flanges were modeled as beam elements rather than
shells. Second, rather than framing cross-frame members into the stiffened girder web (i.e., how
cross-frames are fabricated and erected in practice), cross-frames were framed into the shared
node at the girder web-to-flange juncture to maintain consistency with most 3D bridge-related
software programs. Upon developing the model, the built-in influence-surface feature of Soft-
ware A was utilized to move a 1-kip load along the length and width of the bridge deck, similar
to the procedures outlined previously.
2.5.1.2 2D PEB Models

The 2D PEB models were developed in accordance with AASHTO LRFD Article 4.6.3.3.1.
The concrete deck was explicitly modeled as a shell element. Girders were modeled as beam
elements offset from the deck shell elements to represent the height difference between the
centroids of the respective components. The shells representing the concrete deck and the beam
elements representing the girders were constrained together to simulate composite action.
Given that the depth component of the girders is neglected in the PEB modeling approach,
cross-frames cannot be explicitly modeled. Instead, they are represented as equivalent beams,
as outlined in more detail in Section 2.5.2.
Similar to the 3D approach, bearings were represented as linear springs, except that they acted
at the centroid of the girder section. Additionally, at least one intermediate node was placed on

the beam element representing the girder between two adjacent cross-frame intersections
on that beam. This is particularly important for horizontally curved bridge systems, which
generally model the girders as a series of chorded, straight-line segments (White et al. 2012).
The influence-surface loads were applied directly to the deck shell elements, in the same manner
as the 3D model.
2.5.1.3 2D Grillage Models

For 2D grillage models, the deck is not explicitly modeled with shell elements, which drama
tically impacts the modeling assumptions related to live load application and overall stiffness.
There are, however, procedures to approximately consider the contributions of the concrete
deck as a transverse load distribution mechanism, which are discussed in Section 2.5.3. In terms
of load application in grillage models, live load force distribution is typically based on the simpli-
fied distribution factors specified in AASHTO LRFD. In this case, the lever rule was utilized to
assign a percentage of the load to the neighboring girders.
In terms of stiffness, the girders, much like the 2D PEB approach, were modeled as beam
elements. However, the section properties of the beam elements in grillage models consider the
composite section. Thus, the grillage models inherently reflect the longitudinal stiffness of
the deck but neglect its transverse stiffness. Similar to the PEB approach, at least one intermediate
node was placed along the girder beam element between adjacent cross-frame intersections.
Cross-frames and bearings in grillage models were handled the same way as with the PEB models.
Table 2-4 summarizes the general techniques used by the various analysis methods con-
sidered in the study. The table serves to highlight the major differences in how the key load-
distributing elements were represented, namely the deck, girders, and cross-frames. Note that
Table 2.4 does not consider any modifications that potentially improve the predicted response
of the bridge; however, these concepts are introduced in subsequent sections.
Based on the descriptions of each analysis method, it is apparent that the 2D approaches
rely heavily on simplifications and thus have a greater potential for error. This is particu-
larly true for cross-frame force predictions in 2D models because cross-frames are modeled
as equivalent beams. In the context of Table 2-4, it is expected that the performance of the
analysis method moving from left to right will decrease in accuracy with respect to predicting
cross-frame forces.
2.5.2 Equivalent Beam Approach for Cross-Frames

For both the 2D grillage and PEB modeling approaches discussed in the preceding sub
section, cross-frames are commonly converted into equivalent beams for analysis purposes
Table 2-4. Parameters considered in the analytical-model parametric study.
Analysis Method
Element
Control 3D 2D PEB 2D Grillage
Concrete deck Shells Shells Shells --a
Girders Shells Shells/beams Beam elementsb Beam elementsb
Cross-frames Truss elements Truss elements Equivalent beamsc Equivalent beamsc
Notes:
a
Grillage models do not explicitly consider the concrete deck.
b
In the 2D PEB model, beam elements represent the steel section alone; in the 2D grillage model, beam elements
represent the effective composite section.
c
There are several methods by which the equivalent beam section properties are computed, as discussed in
Section 2.5.2.

and converted back into an idealized truss system for obtaining internal member forces. With
respect to this transformation, there are two major questions that often arise when imple-
menting these procedures into 2D models. The first is related to the section properties of the
equivalent beam elements. Several different methods are commonly used to compute the
equivalent moment of inertia, torsional constant, and shear area as outlined in Section 2.5.3.
The second question is related to the postprocessing procedures once the shear forces and
end moments in the equivalent cross-frame beams are obtained from the 2D models. Typically,
these end moments and shears are applied as external loads on an idealized truss model, from
which internal axial forces in struts and diagonals are computed. The end moments are often
resolved as a force couple between the top and bottom nodes of the truss, resulting in equal-
and-opposite forces in the top and bottom struts. For X-type cross-frames, the end shears are
equally distributed between top and bottom nodes resulting in equal-and-opposite diagonal
member forces. In many cases, these behaviors are seldom observed. For K-type cross-frames,
the vertical shear component is resisted entirely by the single diagonal member framing into a
given end, such that no distribution assumption is required.
Aside from the general procedures, this equivalent beam approach is also thought to produce
cross-frame force results with varying levels of accuracy, particularly for heavily skewed and/or
curved bridges. With the background information outlined, the next subsection offers and
explores specific improvements on these simplified 2D methodologies.
2.5.3 Approaches for Improving 2D Analyses

In conjunction with the 2D models outlined previously, several improvement techniques
based on NCHRP Research Report 725 (White et al. 2012) and past experience were explored in
the Commercial Design Software Study. These improvement techniques range from analysis
to postprocessing assumptions. These techniques, along with their designation as analysis or
postprocessing modifications, are briefly summarized as follows:
1. Improve the equivalent cross-frame beam properties [Analysis]. As outlined in NCHRP
Research Report 725 (White et al. 2012), there are three approaches by which cross-frames
can be transformed into an equivalent beam for use in 2D models: the flexural-analogy
approach, the shear-analogy approach, and the Timoshenko beam approach. Per NCHRP
Research Report 725, the Timoshenko beam approach provides the most realistic estimate
of the cross-frame stiffness because it considers both flexural and shear deformations.
In this study, this assertion is verified in the context of composite systems. For the sake
of brevity, refer to Section 3.2.3 of NCHRP Research Report 725 for a detailed description of
each method.
2. Use an equivalent torsion constant for the beam elements representing girders to account for
the neglected warping stiffness [Analysis]. As documented in NCHRP Research Report 725,
many software programs, including Software A and Software B, neglect the warping stiffness
term when idealizing girders as beam elements in 2D analyses. This results in a significant
underestimation of the torsional stiffness of the system. Significant errors in predicted
bridge response (e.g., deflections and cross-frame forces) have been shown to occur in non-
composite systems, particularly in skewed and horizontally curved bridges.
Given that the current study is focused on composite systems subjected to live loads,
the equivalent torsion constant (represented as Jeq) proposed by White et al. (2012) was
explored, but it was found that this modification is less impactful for composite systems.
In a composite system, the torsional stiffness of the concrete deck is substantial compared to
the added contribution of the warping stiffness in the girders. Therefore, implementing an
equivalent torsional constant for girders was shown to have little effect on the cross-frame

force predictions. For completeness, the models and corresponding results presented later in
this report include equivalent torsional properties.
3. Consider the transverse stiffness of the concrete deck [Analysis]. Given that 2D PEB models
represent the deck as a shell element, this technique is only applicable to grillage models. In
this modification, the transverse stiffness of the deck can be simulated using (i) equivalent
section properties or (ii) notional beams.
For the first approach, the section properties of the equivalent cross-frame beam can be
modified by simply summing the contributions from both the cross-frame (as previously
determined by the Timoshenko beam approach) and the deck, whose effective width is taken
as half the distance to the nearest cross-frame on each side. Thus, the additional moment
of inertia, shear area, and torsional constant contributed by the applicable concrete deck
strip (as modified by the appropriate modular ratio) is added to the equivalent cross-frame
beam properties. In addition to the effective section property approach, consideration of
the transverse deck stiffness can also be implemented with separate, notional transverse
beam elements in the grillage model per AASHTO Article C4.6.3.3.4.
4. Reconsider the distribution of shear forces and moments when postprocessing results [Post
processing]. As introduced previously, the end moments obtained from 2D analyses are often
resolved as equal-and-opposite force couples to the top and bottom nodes of the truss, and
end shears are assumed equally distributed to top and bottom nodes for X-type cross-frames.
These assumptions, although simple to implement, can produce significant errors for com-
posite bridge systems as is shown by the results herein. Three-dimensional FEA results and
measured data have shown that top and bottom struts generally do not have equal-and-
opposite forces; similarly, diagonal members generally do not have equal-and-opposite forces
in X-frames. As such, a unique postprocessing tool is proposed for both 2D grillage and
2D PEB models, as illustrated in Figure 2-17.
The grillage improvement technique (left side of Figure 2-17.) is related to the equivalent
section properties approach outlined in item #3 above (i.e., considering the contributions
of the transverse deck stiffness). By effectively increasing the stiffness of the equivalent beam
to account for the contributions of the concrete deck, considerations for the deck must also be
made when processing 2D analysis results. Otherwise, the force demands on the cross-frame
elements will be artificially amplified due to the increase in assigned stiffness. In contrast,
the methodology below is not applicable to the notional beam approach introduced in
item #3. For that case, the end moments and shears obtained for the equivalent cross-frame
beam can be resolved directly to the cross-frame truss model given that the contributions
of the concrete deck are explicitly and independently considered in analysis.
As such, rather than applying the end moments and shear forces on the cross-frame
panel alone, the applied loads are resolved on a pseudo-composite system including the
cross-frame and deck. The end moment is resolved between the centroid of the deck and
the centroid of the bottom strut, which results in a moment arm of H (or the distance
M1 / H
Vdeck FTS M1 / hb FTS

(1-n)VCF
VCF H
hb
nVCF
M1 / H FBS M1 / hb FBS
Figure 2-17. Postprocessing improvement methods

for 2D grillage models (left) and 2D PEB models (right).

between the respective centroids). This modification is intended to remedy the problem of
having equal-and-opposite strut forces in composite systems. It should be noted that in
noncomposite systems, equal-and-opposite force distribution in the struts is anticipated.
Thus, it is apparent from this simple discussion that the behavior of cross-frames in non
composite and composite systems is different and should be treated as such in the cor
responding analysis. This discussion is advanced further in subsequent sections.
Additionally, in this proposed technique for 2D grillage models, the shear force acting
on the equivalent beam must be distributed to the concrete deck and cross-frame panel.
The research team utilized a rational approach in which the shear force is distributed
based on the relative stiffness of the deck and cross-frame to determine Vdeck (sheer force
component resisted by the concrete deck) and VCF (shear force component resisted by the
cross-frame), although other methodologies are possible. These expressions can be found in
Appendix F. The distribution of VCF to the top and bottom nodes of the cross-frame truss is
discussed below.
The PEB postprocessing technique (right side of Figure 2-17) differs from the grillage
technique in that the internal moments and shears obtained from the analysis (via equiva-
lent beams) only consider the contributions from the cross-frame panel. Recall that the
stiffness of the concrete deck is explicitly considered by shell elements in PEB models.
Thus, in the process of converting end moments and shears into cross-frame force effects,
the added discussion about the moment arm H above is not relevant.
With that in mind, the remaining discussion relates to how the VCF acting on the cross-
frame panel is distributed to the top and bottom nodes. This significantly impacts the
assumed force effects in the cross-frame members, particularly for the diagonals since the
force effects in these members are directly related to the vertical component of the nodal
forces. The research team explored several solutions to this problem and compared the
behavior of noncomposite and composite systems. Ultimately, the equal distribution “50-50”
assumption (i.e., 0.5VCF to the top node and 0.5VCF to the bottom node) is compared to
a “100-0” assumption (i.e., conservatively and independently evaluate the cases in which
100% of VCF is resisted by the top node and 100% of VCF is resisted by the bottom node).
This is demonstrated schematically in Figure 2-17, where the percentage of shear force
distribution is represented by the variable n.
Table 2-5 summarizes the analysis-related improvement techniques outlined above and the
relative impact on deck, girder, and cross-frame elements. Blank table entries indicate that no
modifications are made to that particular element.
Table 2-5. Summary of improved analysis techniques for 2D models.
2D Analysis Improvement Techniques

Element Improve Equivalent Equivalent Torsional Equivalent Cross-Frame
Cross-Frame Beamsa Constant for Girdersa Beamsb
Consider with equivalent
Concrete deck --c -- c
cross-frame beam
Assign Jeq to girder
Girders -- c -- c
properties
Use Timoshenko beam Adjust equivalent section
Cross-frames -- c
approach properties to include deck
Notes:
a
Applies to both grillage and PEB models.
b
Applies to grillage models only.
c
Blank cells indicate that the improvement technique does not directly apply to that structural element.

2.5.4 Parametric Study Overview

To assess the limitations of these various 2D modeling techniques and modifications,
a series of analyses were conducted as part of this Commercial Design Software Study. The
results of 3D truss-element models and various 2D models were compared for a variety of
bridge geometries, where the 3D models served as the basis for comparisons. More specifically,
the research team compared 3D truss-element models developed in Abaqus, which were used
in the Fatigue Loading Study, with the following analysis methods in the general-use software
package, Software A:
• 3D model (cross-frame modeled with truss elements; stiffness modification, R, taken as 0.6
to maintain consistency with Section 2.3.1.1 of this report),
• 2D PEB model (cross-frames idealized as equivalent beams; modifications outlined in
Section 2.5.3 and R = 0.6 inherently considered in equivalent beam properties), and
• 2D grillage or grid model (cross-frames idealized as equivalent beams; modifications
outlined in Section 2.5.3 and R = 0.6 inherently considered in equivalent beam properties).
As noted above, the stiffness modification factor (R = 0.6) was directly applied in the
development of the equivalent beams in the 2D models, thereby reducing the equivalent beam
section properties. The various models outlined above were developed for a scaled version of the
4,104-model matrix outlined in Section 2.3.1.2. Ultimately, the same 20 representative bridges of
the 4,104 total introduced in Section 2.3.3 were evaluated. These representative bridges sampled
key parameters from the full matrix that most often affect cross-frame force predictions in
2D simplified analyses (e.g., support skew and horizontal curvature). Thus, information or
knowledge gained from this abbreviated study is directly applicable to a broader range of
bridges. For each of the 20 representative models, the research team conducted an influence-
surface analysis (or equivalent) for the various iterations. For reference, the pertinent variables
that describe the overall geometry and cross-frame layout of these sample bridges can be
found in Appendix F. The results of the Commercial Design Software Study are subsequently
presented in Section 3.4.
2.6 Stability Study

The Stability Study focused on investigating two major design issues related to Objective (e)
introduced in Section 1.2: (i) development of stability bracing requirements for steel I-girders
extending the available solutions to include negative moment regions and (ii) combination of
stability bracing strength requirements with consideration of force effects generated during
construction and throughout the service life of the bridge. Although the major stability force
effects likely come from gravity loading during deck casting, the impact of forces induced from
other sources such as wind and overhang construction loads needs to be considered. This section
addresses the methodology and background associated with item (i), whereas item (ii) is largely
addressed in Section 3.5.4 and the design examples (Appendices B and C).
As noted previously, AASHTO LRFD currently has no guidance for stability bracing
requirements. As such, the primary goal of this study was to develop design guidance in the
context of steel I-girder bridge systems, using the bracing provisions in the AISC Specifications
as the template. Consequently, the AISC provisions were reviewed and modified accordingly
based on a series of finite element studies introduced in this section. First, a cursory overview of
the AISC guidelines for torsional braces is provided below for context.
Effective stability bracing of beams can be provided by either restraining twist of the cross-
section (torsional brace) or restraining lateral movement of the compression flange (lateral
brace). Cross-frames generally restrain twist of the section and are therefore categorized as

torsional bracing. Regardless of whether a system is categorized as torsional or lateral bracing,

adequate stability bracing must satisfy both stiffness and strength requirements. The stiff-
ness requirement, in general terms, is instituted to limit the twist and/or lateral displacement
of a girder at the brace point. The recommended brace stiffness is often given as an integer
multiplier of the “ideal” brace stiffness (βi), which corresponds to the brace stiffness required
for a perfectly straight member to reach a specified capacity or load level. Since “real” structural
members are not perfectly straight (i.e., they possess some initial imperfection or out-of-
straightness over the length), the current AISC design provisions assume that twice the ideal
stiffness adequately controls deformations and brace forces.
Based on the work conducted by Winter (1960), it was found that providing twice the
ideal stiffness (2βi) limits the out-of-plane deformations for columns to a value equal to the
initial imperfection as the applied load reaches the critical buckling load. This observation,
coupled with simplifications covered in the commentary (Yura 2001), is the basis for the required
torsional brace stiffness adopted in the AISC Specifications (2016):
2.4 LM r2
b T,req = 2.1
fnEI yeff Cb2
where
bT,req = required system torsional brace stiffness,
L = span length,
Mr = maximum factored moment within the critical unbraced segment,
n = number of intermediate braces within the span,
E = the modulus of elasticity of steel (29,000 ksi),
Iyeff = effective moment of inertia,
f = resistance factor, taken as 0.75, and
Cb = moment gradient factor assuming the beam buckles between the brace points.
These variables, in the context of single- and reverse-curvature conditions, are discussed
in Section 3.5.
Although this “twice the ideal stiffness” assumption works well for columns, studies on
beam torsional bracing have shown a larger value may be warranted. In response, the research
team conducted a parametric study that investigated the effects of girder cross-section, loading
conditions, intermediate bracing schemes, girder spacing, and number of girders on the
required stiffness of a torsional beam brace (i.e., a cross-frame). A brief overview of the research
methodology used in this study is provided in Section 2.6.3.
While a reasonable view of stiffness requirements might focus solely on the stiffness of the
cross-frame, in reality, the total torsional brace stiffness is a function of several components.
The total torsional stiffness of a brace is a combination of three main components including:
(i) brace stiffness, b b, (ii) cross-sectional distortion stiffness, bsec, and (iii) in-plane girder stiff-
ness of the beams, bg. The AISC bracing provisions do not include the expression for in-plane
girder stiffness since that component is not a major factor in most building applications.
However, for bridge applications, bg can significantly impact the behavior, particularly for
relatively narrow girder systems (Yura et al. 2008; Han and Helwig 2016). The individual
stiffness components tend to follow the expression for springs in series, as demonstrated
mathematically with the following expression:
1 1 1 1
= + + 2.2
b T b b b sec b g

A review of Eq. 2.2 demonstrates that the total stiffness of the system, b T, is always less than
the smallest of the three individual terms on the right of the equation. For example, a stiff cross-
frame with flexible connection details will severely limit the stiffness response of the system;
or a flexible in-plane girder stiffness, which effectively equates to a narrow superstructure, will
diminish the adequacy of a cross-frame. The previous research related to these topics is sum-
marized by Helwig and Yura (2015). While Eq. 2.2 provides the total stiffness of the system, in
design, the actual brace stiffness must exceed the required stiffness, as determined by Eq. 2.1
(i.e., b T ≥ b T,req).
In addition to stiffness requirements, stability bracing must also satisfy strength require-
ments. In general terms, the required brace strength is a function of the required stiffness and
the assumed initial imperfection. A distinct change in bracing strength requirements, however,
occurred between the 14th Edition AISC Specifications (2010) and the current 15th Edition
(2016). Applying some simplifications which are documented in the commentary, the original
2010 version of the torsional brace strength requirement (based primarily on elastic buckling
behavior) was as follows:
0.024 M r L
M br = b Tq o = 2.3
nCb Lb
where
Mbr = required strength of a torsional brace,
b T = required system torsional brace stiffness (based on twice the ideal stiffness),
qo = initial imperfection in terms of a twist angle, and
Lb = unbraced length of the critical segment.
The remainder of the variables were previously introduced in Eq. 2.1. Note that the initial
imperfection is based on the critical imperfection shape, which for torsional bracing consists
of a lateral sweep of the compression flange equal to Lb/500 (sweep tolerance) while the tension
flange remains straight (Wang and Helwig 2005). This produces an initial twist equal to Lb/500ho,
where Lb and ho are the respective unbraced length and distance between flange centroids.
In contrast, the latest AISC Specification (2016) introduced a change in the torsional brace
moment equation based upon a study conducted by Prado and White (2015). The researchers
carried out a detailed investigation on the stability bracing requirements with an emphasis on
inelastic buckling of relatively short unbraced lengths. This research prompted the revision in
the torsional brace moment equation, which is also given as Eq. A-6-9 in the current AISC
Specifications:
M br = 0.02 M r 2.4
This expression is simply a function of the internal factored moment in the critical unbraced
segment. Although the simplicity of Eq. 2.4 is attractive, the applicability of the expression for
general design situations is questionable when compared to the longstanding strength equations
predicted by Eq. 2.3. To determine which strength design expression is more appropriate for
implementation into AASHTO LRFD, an additional parametric study was conducted in tandem
to NCHRP Project 12-113 (Liu and Helwig 2020).
As noted above, LTB in bridge applications is most critical during girder erection and deck
construction. Although the bending moments during construction are smaller in magnitude
than the live load moments in the completed structure, the noncomposite girders are most
susceptible to instability at this stage. During construction, the steel section alone generally
supports the entire load, and all permanent bracing may not be installed. Additionally,

stay-in-place forms, commonly used to support wet concrete during deck construction, have
connections to the girder that potentially can introduce significant flexibility and therefore are
not considered bracing elements in bridge applications (Egilmez, Helwig and Herman 2016).
While bracing demands are often considered most critical in positive moment regions, both
the positive and negative moment regions in continuous girder systems need to be considered.
In the finished structure, the composite deck provides significant continuous lateral and
torsional restraint to the top flange. For simple spans, for which only the top flange is in
compression, girders are not prone to LTB. In continuous spans, the stability behavior of
continuous girders in the negative moment region (i.e., the bottom flange is in compression) is
often questioned by designers. As a result, cross-frames are often provided in negative moment
regions to control LTB, and the substantial restraint provided by the deck to the bottom flange
is conservatively neglected in design. There are, however, a number of beneficial restraints
that can be considered around interior supports of the finished structure.
As discussed by Yura (2001), the composite deck not only continuously braces the top
flange but also provides additional bracing benefits to the bottom compression flange in the
negative moment regions, assuming web distortion is prevented. Additionally, the girder
bearings themselves provide lateral restraint and some torsional restraint to the bottom
compression flange in these regions, which further mitigates an LTB problem. Therefore, the
stability bracing requirements for cross-frames in negative moment regions of composite
systems are not critical for design.
To demonstrate the beneficial effects of continuous top flange restraint, Figure 2-18
schematically depicts the results of an eigenvalue FEA buckling analysis on a bridge girder
with stiffened webs subjected to reverse-curvature bending, which is representative of many
continuous systems near the interior supports. In all three cases presented, twist of the cross-
section is prevented at the girder ends. However, the level of top flange restraint is modified in
each successive case.
From the buckled shapes, it is evident that cases (i) and (ii), which are representative of the
steel section alone and a noncomposite girder condition (i.e., a concrete deck without shear
studs), respectively, tend to buckle in a traditional LTB (or similar) mode. In these cases, the
unrestrained compression flange(s) displaces laterally in the out-of-plane direction along the
(i) No continuous top (ii) Continuous top (iii) Continuous top

flange restraint flange lateral restraint flange lateral &
torsional restraint
Figure 2-18. Typical buckled shape for beams subjected

to reverse-curvature bending with various degrees of
top flange restraint.

full unbraced length. In case (iii), the buckled shape is indicative of a distortional mode in the
web, which is not sensitive to the unbraced length but is sensitive to the web slenderness and
transverse stiffener details. The buckling capacity of these cases generally increases from left to
right, as presented in Figure 2-18. More specifically, the bucking capacity associated with case
(iii) is typically substantially larger provided that transverse stiffeners are included and are
properly detailed. As such, girder segments in the negative moment regions are often controlled
by yielding or the distortional buckling mode, for which bracing demands in the cross-frames
are much less significant than the construction condition. This phenomenon is supported with
additional analyses outlined in Section 2.6.1.
The critical stage, therefore, occurs during erection and deck construction with the non-
composite steel girder system supporting the entire construction load. With that in mind, the
studies outlined herein are focused primarily on noncomposite systems under various moment
gradients including single- and reverse-curvature bending, as these conditions represent the
most critical in terms of bracing requirements.
Before outlining the stability bracing studies, a cursory overview of a computational study
investigating buckling in composite conditions is provided in Section 2.6.1. An overview of the
parametric studies conducted to address these bracing strength and stiffness requirements are
then provided in the following subsections. Section 2.6.2 outlines the bracing strength study,
and Section 2.6.3 outlines the bracing stiffness study.
2.6.1 Buckling in the Composite Condition

To verify the assertion that buckling in the composite condition is not critical for design,
several spot check computational studies were performed in the general purpose FEA software,
Abaqus. Elastic eigenvalue buckling analyses were conducted on prismatic girder segments
with unbraced length-to-depth ratios (i.e., Lb /d) of 5 and 10 and a slender flange-width-to-
web-depth ratio (i.e., bf /d) of 1/6—the minimum value permitted in AASHTO LRFD. Cases
with closely spaced transverse web stiffeners (i.e., stiffener spacing equivalent to the girder
depth) and no transverse stiffeners along the length were also studied to examine the effects of
these details on web distortion buckling.
Rather than investigating entire girder systems with bracing elements as is done in the
subsequent sections, the critical unbraced segment in the negative moment region is isolated
and analyzed independently, while conservatively neglecting the additional warping restraint
provided by the adjacent unbraced segments in the span. Moment gradients along the length
of the critical segment are then simulated through a series of end moments. This procedure
is consistent with past LTB studies that investigated concepts such as load-height effects and
moment gradient factors for singly symmetric and nonprismatic sections (Helwig, Frank and
Yura 1997; Reichenbach et al. 2020).
With that in mind, various straight-line moment gradients were considered with varying
degrees of negative flexure in the unbraced segment, ranging from uniform negative moment
(i.e., constant compressive stress in the bottom flange) to equal-and-opposite end moments
(i.e., straight-line moment diagram with an inflection point at mid-length of the segment).
These parameters were selected to represent practical conditions commonly found in continuous,
composite bridge girders near the interior supports.
The primary intent of these studies was to evaluate the beneficial effects of continuous top
flange restraint in the negative moment regions of composite systems. Thus, in addition to
the torsional restraint provided by the cross-frames at the ends of the critical girder segments,
three different types of additional top flange restraints were considered: (i) no additional top

flange restraint, which is representative of the steel section alone during deck construction,
(ii) continuous lateral restraint to the top flange, which is representative of noncomposite
finished bridges not utilizing shear connectors, and (iii) continuous lateral and torsional restraint
to the top flange, which is representative of a composite bridge in its finished state. Case
(iii) represents an upper-bound condition in terms of restraint fixity, as the composite deck
does not supply a completely rigid torsional restraint to the top flange.
For the various parameters and moment gradients, the critical buckling moment was
evaluated and compared as a function of the top flange restraint provided. The critical eigen
vector either corresponded to an LTB mode or a web-distortional buckling mode, similar to
the illustrations provided in Figure 2-18. The results of this study are subsequently presented in
Section 3.5.1.
2.6.2 Bracing Strength Study

To investigate torsional bracing strength behavior, a study was conducted in tandem with
the NCHRP Project 12-113 study consisting of parametric 3D finite element analyses on twin
I-girder systems using Abaqus (Liu and Helwig 2020). The results are directly applicable to
the NCHRP Project 12-113 study, and therefore the following section provides a cursory over-
view of these studies. For a more detailed outline of the modeling assumptions and procedures,
refer to Appendix F.
In general, a series of buckling analyses on twin-girder systems with an assumed initial
imperfection were conducted. Two girders represent the fewest number of girders that are
possible with torsional bracing provided by cross-frames or diaphragms; the bracing behavior
is representative of systems with more than two girders. A variety of intermediate bracing
configurations were considered, including one to five cross-frames between the girder supports.
Both elastic and inelastic (i.e., assuming elastic-perfectly plastic behavior with yield strengths
of 36, 50, and 70 ksi) material analyses were carried out to examine the influence of material
nonlinearity on the response of the girders and cross-frames. In general, tailoring brace strength
requirements around a specific material yield strength is not advisable, as it can potentially
lead to unconservative estimates of the stability brace moments. Instead, provisions for brace
strength should be applicable for a variety of yield strengths. Strain hardening and residual
stresses, however, were not considered in the study since preliminary results demonstrated that
elastic materials produced more critical results.
Three different noncomposite, prismatic girder cross-sections (referred to as Cross-section 1,
2, and 3 herein) were examined, which primarily investigated different flange-width-to-web-
depth ratios (bf /d) ranging from the extreme limit for built-up sections permitted in AASHTO
LRFD (2020) to a value more consistent with typical rolled sections. These cross-sections are
depicted schematically in Figure 2-19. The web depth and span length were subsequently held
constant to achieve a span-to-depth ratio (L/d) of 25, which is representative of common
bridge girders used in practice.
In addition to various girder cross-sections, the research team also examined different
loading conditions, including uniform moment and uniformly distributed loading (i.e., single-
curvature bending). The distributed loads were either applied at the top flange or at mid-height
of the sections to study the impact of load position on the cross-section. As part of NCHRP
Project 12-113, additional spot checks were performed to examine bracing demands in
bridge girders subjected to reverse-curvature bending, particularly for cross-frames in the
negative moment region. As noted previously, only the noncomposite condition is repre-
sented in these various loading conditions. Thus, no lateral and torsional restraint provided
by a composite concrete deck is considered. For all loading scenarios, the girders were simply

Flanges:
8"×0.5" 12"×0.75" 16"×1"
0.75"
48"
Cross-section 1 Cross-section 2 Cross-section 3

Extreme Limit Typical Built-Up Typical of Rolled
in AASHTO Shapes W-Shapes
Figure 2-19. Different cross-sections

evaluated in the stability bracing
parametric studies.
supported at the ends. For the cases involving reverse curvature, the girders were still simply
supported, but girder continuity was simulated with applied end moments (positive and
negative). Twist was also restrained at the girder ends, but the sections were free to warp.
For the single-curvature bending cases, it was observed that the critical segment for buckling
that consistently resulted in the largest brace forces was near midspan, where positive moment
and out-of-plane girder displacements were maximized. As such, a critical asymmetric imper-
fection consistent with Prado and White (2015) (i.e., the critical compression flange is dis-
placed laterally in accordance with the discussion in the preceding section) was assumed in these
critical areas.
For reverse-curvature bending, it is not always as clear which segment is critical for buckling.
In continuous girder systems, the negative moment regions generally have the largest moment
magnitudes. However, these regions are aided by steeper moment gradients (i.e., larger Cb factors)
and restraint provided by the nearby girder supports. In contrast, positive moment regions have
smaller moment magnitudes but typically have Cb values close to unity, especially as additional
intermediate braces are included. For example, in the interior span of a three-span continuous
unit, the maximum negative moment is approximately double the maximum positive moment.
However, the moment gradient factor in the negative moment region is often greater than 2.0
(depending on the unbraced length) based upon published values or AISC expressions, compared
to a Cb factor near unity for the positive moment region.
Thus, it was observed that the critical brace (i.e., maximum bracing forces) depended on
the location of the critical imperfection. For instance, the midspan brace forces in the positive
moment region were maximized when the imperfection was assumed along the compression top
flange at that same location. In contrast, the brace forces at the first intermediate brace line in the
negative moment region were maximized when the imperfection was assumed along the bottom
(compression) flange at that location. Note that the imperfection in the negative moment case
was not implemented directly at the support condition, where negative moment is actually
largest, due to the presence of bearings that benefits the torsional bracing requirements.
Permutations of different girder cross-sections, bracing configurations, and loading conditions
were systematically analyzed by a series of independent eigenvalue and incremental analyses.
Eigenvalue analyses were initially performed to obtain the ideal stiffness of the cross-frame
required to buckle the girders between the brace points. The large-displacement incremental
analysis, performed on an imperfect system and cross-frames with twice the predetermined

ideal stiffness, was subsequently performed to obtain the relationship between internal girder
moments and critical bracing moments. An additional discussion on modeling convergence
and inferring cross-frame brace forces from the results is provided in Appendix F for reference.
The results of this stability bracing strength study are subsequently presented in Section 3.5.2.
2.6.3 Bracing Stiffness Study

To investigate the torsional brace stiffness behavior, a similar parametric 3D FEA study
was carried out in Abaqus. Many of the same features and modeling assumptions outlined
in Section 2.6.1 were also examined in this study, including the various intermediate bracing
configurations, girder cross-section proportions, load-height effects, and critical imperfections.
There were two notable differences, though, between the brace strength study outlined previ-
ously and the brace stiffness study.
First, the effect of material inelasticity was not explicitly considered in the stiffness study.
As noted in Section 2.6.1, it was demonstrated that elastic materials generally produce more
critical demands for cross-frame stiffness and strength. As such, only elastic materials with no
residual stresses were considered in these FEA studies. Second, the bracing stiffness study exam-
ined redundant bridge systems, including those with two, three, four, and six girders across
the width. Regardless of the number of girders, the applied loads and critical imperfections
outlined above still applied.
These variables were parametrically evaluated through a series of independent eigenvalue and
incremental analyses. Eigenvalue analyses, as outlined in Section 2.6.1, were initially performed
to obtain the ideal stiffness of the cross-frame required to buckle the girders between the brace
points. An incremental analysis on the imperfect system was subsequently conducted for a
variety of cases. More specifically, the stiffness of the cross-frames was examined for its effect
on the out-of-plane girder twists at the brace points (relative to the initial imperfection) for
increased load levels. To examine these effects, brace stiffness to ideal stiffness ratios, b b/b i,
of {2, 2.5, 3, and 4} were studied. For each stiffness multiplier, girder twists were obtained as
a function of increasing internal girder moments. Ultimately, the objective was to determine
which ideal stiffness multiplier produced final girder twists equal in magnitude to the initial
imperfection. Additional discussion on the overall methodology and procedures is provided
in Appendix F. Results of this stability bracing stiffness study are summarized Section 3.5.3.

CHAPTER 3
Findings and Applications
As noted in Chapters 1 and 2, the five major objectives of NCHRP Project 12-113 are
addressed with both experimental and analytical studies. The experimental portions included
the instrumentation and monitoring of three bridges, while the analytical studies included four
independent studies (Fatigue Loading Study, R-Factor Study, Commercial Design Software
Study, and Stability Study). While Chapter 2 outlined the general means and methods for each
study, this chapter summarizes the key findings and results of those studies.
Similar to the previous chapters, the data presented herein is not intended to provide a
comprehensive overview of the experimental and analytical results. Rather, sample results are
provided that highlight the key observations to contextualize the conclusions and proposed
modifications to AASHTO LRFD presented in Chapter 4. For a more detailed overview of the
results, refer to Appendices E and F.
This chapter is divided into five major sections similar to the organization of Chapter 2.
Section 3.1 outlines the major results from the field experimental program. The controlled live
load test and the model validation studies are summarized, followed by the in-service stress
data that supplements the computational results. Section 3.2 presents the key results from the
Fatigue Loading Study with a particular emphasis on AASHTO fatigue loading criteria and
WIM records. Section 3.3 summarizes the investigation of the eccentric connections for cross-
frames and the proposed R-factors. More specifically, the reliability of the stiffness modification
approach for 3D modeling of cross-frames is explored, as well as a proposed eccentric-beam
approach. Section 3.4 reviews the data related to the Commercial Design Software Study, where
the limitations of various 2D modeling methods are examined. Lastly, Section 3.5 provides
a summary of the major findings related to the Stability Study. Note that the results of the
industry survey were previously discussed in Section 2.1, so no additional commentary is pro-
vided in this chapter.
3.1 Field Experimental Program

and Model Validation
As outlined in Section 2.2, two different types of field experiments were conducted in Phase II
of the project for three different bridges (a straight bridge with normal supports, a straight
bridge with skewed supports, and a horizontally curved bridge). The controlled live load test,
for which trucks of known axle configuration and weight were statically positioned at different
locations along the deck width and length, provided valuable data to aid in the validation of an
FEA modeling approach. The in-service monitoring study, which measured stress cycle counts
in instrumented cross-frame members and girder flanges subjected to realistic traffic conditions,
provided context to the computational results obtained from the Fatigue Loading Study.
60

Findings and Applications 61
Given the distinct differences in the two experimental tests, this section of the report is divided
into two major subsections. Section 3.1.1 highlights key findings from the controlled live load
tests, and Section 3.1.2 summarizes the measured rainflow-counting data considering one month
of measured traffic data on the instrumented components of the bridges.
3.1.1 Controlled Live Load Test and Model Validation

As noted in Section 2.2.2.1 the controlled live load tests performed on each bridge consisted of
seven static load cases as well as a series of moving load cases. The static load cases were primarily
conducted to obtain clean and concise data by which 3D FEA models could be validated. The
moving-load cases were conducted at slow speeds to obtain general influence-line measure-
ments for the instrumented components as well as load-position sensitivity (i.e., how truck
placement impacts cross-frame force response). Before addressing the model validation results,
sample measured influence lines will first be examined to introduce load-position effects.
Figure 3-1 presents the axial-stress influence-line plot for select instrumented cross-frame
members near the maximum positive dead load moment region in the straight bridge with
normal supports that is referred to as Bridge 1. The geometry and cross-frame layout of Bridge 1
were previously summarized in Section 2.2.1. For additional reference, a plan and cross-section
view of the instrumented span (i.e., 194-foot end span of three-span continuous unit) is pro-
vided in Figure 3-2. Note that the identification system used for cross-frames in this figure differs
from what is presented in Appendix E to simplify discussions in the report. The remainder of
Line 4
Brg.
Brg.
CF
2.40
1.60 TS2
D2-2 D2-1
0.80 BS2
TS2
0.00
-0.80
2.40
1.60
Axial Stress (ksi)
D2-2
BS2
0.80
TS2
0.00
D2-1
-0.80
2.40
1.60
0.80
D2-1
BS2
0.00
TS2
D2-2
-0.80
0 50 100 150 200
Distance from Start of Bridge (ft)
Figure 3-1. Influence-line plots for various instrumented

cross-frame members.

Cross-frame with gages
Cross-frame gages
Bottom flange gage
TS2
D4-2 D4-1 D3-2 D3-1 D2-2 D2-1 D1-1
D1-2
BS3 BS2
Figure 3-2. Plan and cross-section views (line 4) illustrating the

instrumentation locations and static load case that correspond to
the sample results in this section.
the figures presented in this section, both the static and moving-load cases, are based on instru-
mented cross-frames in Bridge 1 unless noted otherwise. The full set of field data related to all
three bridges can be found in Appendix E.
Three distinct loading scenarios are presented sequentially in the figure. In the first load
case (top plot), a three-axle dump truck (with an approximate gross vehicle weight of 50 kips)
slowly traversed the full length of the three-span continuous unit along the inside edge of the
left bridge barrier. Similar dump trucks subsequently traversed the full length of the bridge at
different transverse lane positions. For each scenario, the axial-stress time-history in the cross-
frame members of interest was recorded by the DAQ system.
The time-history responses were then converted into influence-line plots by aligning the
time component of the measured data with the longitudinal position of the truck relative to the
start of the span. This is demonstrated in the horizontal axis of Figure 3-1. Additional bench-
mark distances, such as cross-frame (CF) line 4 where the instrumented members are located,
and bridge supports (abbreviated as “Brg.” in the figures for “bearings”) are included for
reference. Note that only the instrumented 194-foot end span of the three-span continuous unit
is plotted along the horizontal x-axis. The influence of applied load on the adjacent two spans
with respect to cross-frame force effects was negligible.
It is also important to note that the data obtained from the strain gages illustrated in this fig-
ure simply provide the change in strain/stress during the applied loading. The data does not
indicate the state of stress prior to the gage being installed. As such, the measured response
does not provide information on the permanent state of stress due to dead loads or residual
stresses.
In review of the sample influence-line results provided in Figure 3-1, there are several key
observations:
• Load-induced cross-frame response is highly sensitive to the transverse position of the
truck. For instance, the cross-frame diagonal designated as “D2-2” experiences a tensile

stress cycle when the truck traverses along the left-hand side of the panel but experiences a
compressive stress cycle when the truck transverses along the right-hand side.
• The influence of longitudinal load position is localized, as demonstrated by the fact that the
measured axial stresses are nearly zero when the truck is positioned beyond 50 feet from
the panel of interest, regardless of the lane position.
• The various cross-frame members in the instrumented panel also have highly varied
responses to a given load condition. In general, top strut (TS) force effects are negligible,
which is attributed to the composite nature of the superstructure. Diagonal and bottom strut
(BS) forces are generally more substantial, but the stress magnitude and sign are dependent
on truck position and the corresponding load-induced deformation pattern of the cross-
frame panel.
• Although not explicitly shown in the figure, similar behavior was observed from the results
for the straight bridge with skewed supports (Bridge 2) and the horizontally curved bridge
with radial supports (Bridge 3). The overall load path and load-induced response of the cross-
frames in those more complex framing systems, however, produces more interesting results.
The effects of skewed supports and horizontal curvature are examined computationally in
Section 3.2.
In terms of the static load cases performed during the field experimental program, four
different dump trucks were incrementally positioned on the bridge deck one at a time. Thus,
for each individual load case, the result was a stepped time-history response of increasing
stress. An example of this is presented in Figure 3-3, where measured bottom flange stresses
in Girder 1 (measured at cross-frame line 4 in accordance with Figure 3-2) are graphed as a
function of time for a specific static load case. There are six distinct steps that are apparent
in the time-history plot, as follows:
A. No trucks positioned on bridge; no live load-induced stress.
B. Truck 1 (lead truck) positioned on the bridge; live load-induced stress increases.
C. Truck 2 positioned on the bridge behind Truck 1; live load-induced stress increases.
D. Truck 3 positioned on the bridge behind Trucks 1 and 2; live load-induced stress increases.
E. Truck 4 positioned on the bridge behind Trucks 1, 2, and 3; live load-induced stress increases.
F. All four trucks are removed from bridge simultaneously; live load-induced stress returns
to zero.
Due to the relatively static nature of this test, the stress response for each instrumented
element essentially follows a step function. Small spikes were periodically recorded because the
7.0
6.0 E
D
Bottom Flange Stress (ksi)
5.0
4.0 C
3.0
2.0 B
1.0
A F
0.0
-1.0
0 200 400 600 800 1000 1200 1400
Time (sec)
Figure 3-3. Measured bottom flange stress for a representative

static load case.

load was not purely static. For example, a small spike was recorded at the beginning of plateau B
in Figure 3-3. This spike occurs as the truck passes over the maximum point on the influence
line for the girder bottom flange, which happens to not coincide with the final static position of
Truck 1 in this sample load case. Had the final position of Truck 1 coincided with the maximum
point on the influence-line plot, the spike would not have been measured. There are also a few
spikes as the trucks are moved off the bridge, which can be explained the same way as the spike in
plateau B. Given that these load cases were performed at very low speeds, the research team does
not believe these spikes are related to a dynamic impact effect of the truck entering the bridge.
The magnitudes of the plateaus are the major focus of the model validation process. These
values indicate the stress imposed on the instrumented elements under static loading condi-
tions. In order to cancel out potential effects of electromechanical noise, an average stress
value, which served as the metric by which the FEA models were validated, was obtained from
each of these load-step plateaus.
An example of static loading results is presented in Figure 3-4. The measured axial stresses
in the instrumented cross-frame members along line 4 are graphically depicted for each
intermediate stage of a load case. The positioning of the trucks relative to the cross-frame line
of interest is depicted in Figure 3-2 for reference. From these results, there are a few notable
observations with regard to the cross-frame response:
• Similar to Figure 3-3, the single instrumented top strut member developed very little axial
force, even for the final stage of the load case that represented over 200 kips of GVW.
0.09
1 Truck
0.47 0.43
0.43
2 Trucks
1.42 1.30
0.59
3 Trucks
2.01 1.80
0.65
4 Trucks
2.24 1.99
Figure 3-4. States of axial stress measured in instrumented

cross-frame during a static load case (Units: ksi).

• Interestingly, all but two instrumented cross-frame members experienced a net tensile stress
under the applied load. This behavior seems to indicate that the simplified postprocessing
methods commonly utilized in 2D analysis programs as outlined in Section 2.5, do not accu-
rately represent realistic conditions. In other words, the diagonal members and strut members
do not have equal-and-opposite force effects under a given applied load.
Cross-frame stresses (e.g., Figure 3-4), girder flange stresses, and girder deflections were
compiled from every static load test performed at all three instrumented bridges. This measured
data served as the metric by which the FEA modeling approach utilized throughout the Phase III
computational studies was validated.
Before the experimental program was executed, preliminary 3D FEA models were developed
based largely on the properties and dimensions specified in the design plans as well as con-
servative design code approaches. In general, these models consistently overpredicted girder
stresses, deflections, and cross-frame forces. Using the measured data as a benchmark, the
research team was able to adjust several of the key modeling assumptions. Ultimately, it was
shown that three modeling parameters, when adjusted, had the most impact in achieving
good agreement with the measured results. These three parameters, which are briefly intro-
duced herein, included boundary conditions, contribution of concrete barriers, and the elastic
modulus of the concrete deck. For a more detailed review of these modeling parameters, refer
to Appendix E.
In short, the boundary conditions were modified from pure pinned and roller conditions
to a more accurate linear spring representation of the bridge bearings. The stiffness of the
elastomeric bridge bearings was estimated based on AASHTO LRFD Equation 14.6.3.1-2
and the dimensions and parameters specified on the design plans. The stiffness of the bridge
bearings was most influential for instrumented Bridge 3 (the horizontally curved system), which
is intuitive given the torsional response of the horizontally curved superstructure to vertically
applied loads.
Conventional analysis and design practices generally neglect the contributions of a bridge
barrier, continuous or discontinuous. The preliminary analyses, much like the conventional
approach, also neglected the barriers in the 3D model. After several model iterations, it was
evident that including the discontinuous bridge barriers markedly improved the cross-frame
force predictions in the FEA models, particularly for Bridge 3, which is a narrow system. By
providing another load path for truck loads, barriers reduce the force demands in the cross-
frame elements. It is acknowledged, however, that including the barriers in analysis models
increases the computational efforts by both designers and programmers. Thus, the conventional
approach of not modeling barriers typically results in a conservative estimate of cross-frame
force effects.
Lastly, the assumed modulus of elasticity of the concrete deck and rails was also shown
to significantly affect girder and cross-frame response in 3D FEA models. In the preliminary
analyses, the research team based the assumed elastic modulus on the minimum concrete
strength specified by the design plans and the traditional American Concrete Institute (ACI)
equation (i.e., fc′ where f ′c is the concrete strength in psi). It was recognized, however, that
design codes inherently provide lower-bound estimates. As such, the research team explored
different publications and experimental data to justify a higher elastic modulus for concrete
(i.e., a value more representative of mean conditions as opposed to lower-bound conditions)
(Tadros et al. 2003), which in turn improved comparisons between the analytical solutions
produced by the models and the experimentally measured data. In design practice, making
these assumptions would be challenging given the uncertainty in the material properties. Thus,
similar to the bridge barrier discussion above, the conventional approach of using code-based
material properties typically results in a conservative estimate of cross-frame force demands.

Table 3-1. Sample Bridge 1 results (girder deflections) demonstrating

the comparisons between preliminary and validated analysis results
with measured data.
Measured Preliminary Analysis Validated Analysis

Element ID
Deflection (in) Deflection (in) Percent Error Deflection (in) Percent Error
Girder 1 0.68 0.75 10% 0.66 –4%
Girder 2 0.67 0.81 21% 0.71 6%
Girder 3 0.70 0.85 23% 0.75 8%
Girder 4 0.79 0.85 7% 0.74 –6%
Girder 5 0.70 0.82 18% 0.71 2%
It is also important to note that in general terms, a user can manipulate a model in many
ways to achieve the target solution; however, those changes may not be a good representation
of the actual structural system. The team was interested in achieving good agreement between
measurements and FEA predictions but not at the expense of using unreasonable assumptions.
Based on a literature review and examining many different model configurations, the research
team was able to select a consistent set of parameters and assumptions that not only improved
the accuracy of the results compared to the measured data, but also made sense based on reason-
able engineering assumptions.
Based on the commentary above, the modeling approach was fine-tuned to improve the
agreement between the measured and finite element results. To demonstrate these improve-
ments for instrumented Bridge 1, Table 3-1, Table 3-2, and Table 3-3 compare the preliminary
and validated analytical results with the measured data for critical girder deflections, girder
stress, and cross-frame force effects, respectively. Only the load case illustrated in Figure 3-2
is presented in these tables; however, the results are representative of all load cases performed
at Bridge 1. For reference, the percent error (relative to the measured data) associated with the
model is also presented.
From these tables, it is apparent that increasing the stiffness of the concrete deck and
including a discontinuous concrete rail stiffened the bridge overall and consistently improved
the analytical results. Critical girder deflections, which were once uniformly overestimated,
improved to errors typically within 10%. Critical girder stresses were also consistently
improved to within 10% to 20% of the measured data. Given the complexity of the bridge
model and the potential uncertainty in the field measurements, these validated discrepancies
are deemed acceptable.
In general, Table 3-3 also demonstrates the improved accuracy of the model with respect
to cross-frame forces. It is important to note that the error associated with cross-frame forces
Table 3-2. Sample Bridge 1 results (girder flange longitudinal stresses)

demonstrating the comparisons between preliminary and validated analysis
results with measured data.

Element ID
Stress (ksi) Stress (ksi) Percent Error Stress (ksi) Percent Error
Girder 1 2.27 2.77 22% 2.73 20%
Girder 2 2.49 3.04 22% 2.91 17%
Girder 3 2.77 3.25 17% 3.10 12%
Girder 4 2.53 3.19 26% 3.02 20%
Girder 5 2.48 3.03 22% 2.94 18%

Table 3-3. Sample Bridge 1 results (cross-frame axial stresses) demonstrating

the comparisons between preliminary and validated analysis results with
measured data.

Element ID
Stress (ksi) Stress (ksi) Percent Error Stress (ksi) Percent Error
D1-1 –2.75 –3.37 22% –3.33 21%
D1-2 5.03 5.92 18% 5.53 10%
D2-1 1.18 –0.69 –159% –0.38 –132%
D2-2 7.67 9.77 27% 8.77 14%
D3-1 7.45 9.07 22% 8.14 9%
D3-2 1.47 0.44 –70% 0.63 –57%
D4-1 5.37 6.66 24% 6.19 15%
D4-2 –2.78 –3.85 39% –3.77 35%
TS2 1.85 9.67 423% 2.94 59%
BS2 5.68 9.67 70% 8.99 58%
BS3 6.41 10.24 60% 9.50 48%
is noticeably higher than what is observed for girder stresses or deflections. Even for the most
sophisticated full-shell, 3D FEA model and a relatively simple bridge geometry, the errors
associated with the critical cross-frame forces still ranged from 0 to 60%. Larger discrepancies
were tabulated, but those correspond to less critical cross-frame members.
Load paths and flexural behavior of girders are generally more straightforward than that
of cross-frames. Furthermore, it should be noted that many of the stress magnitudes were
relatively small such that any slight variations can produce very large percent differences
between measured and predicted stresses. In general, these tables highlight the difficulty with
trying to improve all of the measurements (i.e., girder deflections, girder stresses, and cross-frame
stresses). The system is highly indeterminant when the various components are considered—
multiple girders, many cross-frames, and variations in the concrete deck thickness along the
length and width. Given the complexity of the bridge model and the potential uncertainty in
the field measurements, these validated discrepancies were deemed acceptable.
As noted above, the data presented in Table 3-1 through Table 3-3 represent just one static
load considered at Bridge 1. Similar results were synthesized for every live load case at every
instrumented bridge, which is summarized in Appendix E. By achieving good agreement
between the measured and finite element results in this validation study, it ensured that the
parametric studies executed in Phase III produced reliable results that were consistent with real
load-induced behavior of cross-frames.
3.1.2 In-Service Monitoring

At each instrumented bridge, stress cycle spectra were obtained for various cross-frame
members and girder flanges during a one-month monitoring period. As noted in Section 2.2.2.2,
these spectra provide useful insight on the stress cycle magnitudes that a component typically
experiences due to live loads. The relative difference in truck traffic volume between different
bridges can also be inferred from the data. Thus, the fatigue damage accumulated on the
instrumented bridge components caused by the truck traffic—a function of stress magnitudes
and cycle counts—can be deduced. The spectra, however, are limited in that they provide no
indication of the load spectrum (i.e., the weight and axle configuration of the trucks causing
those cycles) or the corresponding transverse lane positions, which is especially critical for
cross-frames given their observed sensitivity to load position.

The primary goal for obtaining these data was to establish effective and maximum stress
range metrics by which the computational studies and current AASHTO fatigue criteria could
be assessed. Effective stress ranges are related to finite-life behavior (Fatigue II limit state), and
maximum stress ranges are related to infinite-life behavior (Fatigue I). As such, this section
presents representative response spectra data and summarizes the measured effective and
maximum stress range values obtained from the field studies. These metrics are subsequently
used for comparison in later sections of the report.
Measured response spectra are best illustrated as histogram plots, for which the number
of stress cycle counts, determined by rainflow-counting algorithms, are compiled and sorted
into different stress magnitude bins, Sr,j (i.e., jth stress range bin). The effective stress range then
mathematically represents the response of the bridge component to the entire truck population
by equating the fatigue damage caused by the variable-amplitude response spectrum to a
constant amplitude spectrum of equal cycle count. The maximum stress range represents the
upper tail of the spectra. A more detailed discussion on the calculation of effective and maxi-
mum stress range of the spectrum, as well as the truncation process and bin size parameters,
is provided in Appendix E.
A sample histogram illustrating the variable-amplitude stress range spectrum for a cross-
frame member (D1-2) in instrumented Bridge 1 is presented in Figure 3-5 (left graph). The right
graph provides the mathematical representation of the effective stress. That is, 2,683 cycles at
0.91 ksi produces fatigue damage equivalent to the spectra of stress magnitudes and cycle counts
shown on the left graph. The maximum stress range is also approximately 2.7 ksi based on the
upper tail of the spectrum on the left. For clarity, the vertical axis is presented using a log scale.
Aside from the calculation of effective and maximum stress range metrics, there are several
additional observations from Figure 3-5:
• This instrumented cross-frame member in Bridge 1 experienced load-induced stress magni-
tudes ranging from 0 to 2.7 ksi over the entire monitoring period. Note that the stress cycles
below 0.65 ksi were truncated based on the discussion provided in Appendix E, which is
consistent with the method outlined in Connor and Fisher (2006).
• The vast majority of the stress cycles are well below the CAFL for a Category E′ detail (i.e.,
2.6 ksi). In fact, 90% of the recorded stress cycles corresponded to magnitude less than 1.1 ksi.
100,000
Stress Range Spectrum, Sr, j
Total Cycles = 2,683
10,000
1,000
Number of Cycles
2,683 cycles at Sre = 0.91 ksi
100 100
10
0
0 1 2 3 0 1 2 3
Stress Range, Sr, j (ksi) Stress Range, Sr, j (ksi)
Figure 3-5. Graphical depiction of converting variable stress range

spectrum to single effective stress range (cross-frame member in
Bridge 1).

Similar histograms were produced for girder flanges as well. Figure 3-6 presents a side-
by-side comparison of the stress range spectra measured for a cross-frame member (D1-2 in
Figure 3-2) and girder flange (Girder 2 in Figure 3-2) for instrumented Bridge 1. Both instru-
mented elements were subjected to the same load spectrum (i.e., same truck population and
lane positions); however, the response of each is significantly different. Not only are the stress
range magnitudes higher for the girder flanges, but the number of cycles above the correspond-
ing truncation stress was significantly more. Given that the cross-frames were designed for a
much more stringent fatigue detail (i.e., lower CAFL value), it is intuitive that the stress magni-
tudes in girder flanges exceed those in cross-frames. The variation in stress cycle counts, on
the other hand, can be attributed to the sensitivity of cross-frames to transverse load position.
Depending on the precise transverse position of a passing truck, the stress range response in
a particular cross-frame element could be significant or negligible (i.e., below the established
truncation stress or even hidden by the electromagnetic noise of the strain gage). Bending
stresses in girders, however, are less influenced by lane position, as evidenced by the sample
data in Figure 3-6.
Given that (i) the current AASHTO LRFD fatigue load factors were only calibrated for
girder response to truck traffic and (ii) transverse load distribution effects were not explicitly
considered in that study, this measured data gives an initial indication that load cases specific
to cross-frame fatigue are perhaps warranted. This behavior is explored in greater depth
computationally in Section 3-3.
As noted above, the research team produced and processed a number of histogram plots
similar to those presented in this report. In order to evaluate the measured in-service data with
AASHTO fatigue design criteria in subsequent sections, only the effective and maximum stress
ranges are examined. As such, Table 3-4 summarizes the range of effective and maximum stress
ranges computed for the instrumented cross-frame members in Bridges 1, 2, and 3. In other
words, these stress range metrics were computed for every instrumented cross-frame member,
and the range presented in the table represents the maximum and minimum values of the data
set. A similar table with respect to girder flange data is provided in Appendix E for reference.
From Table 3-4, it is obvious that critical cross-frames in Bridge 2 (skewed) consistently
experienced higher effective stress ranges in comparison to critical cross-frames in the other
100,000 100,000
Cross-Frame Girder Flange
10,000 10,000
1,000 1,000
Number of Cycles
Number of Cycles
100 100
10 10
1 1
0 0
0 1 2 3 4 5 0 1 2 3 4 5
Stress Range, Sr, j (ksi)
Figure 3-6. Sample variable stress range spectrum of (left)

cross-frame member and (right) girder flange (Bridge 1).

Table 3-4. Ranges of effective stress ranges and maximum

stress ranges measured for cross-frames at each instrumented
bridge.
Measured Range {Min, Max}

Bridge No.
Effective Stress, (ksi) Maximum Stress, (ksi)
1 {0.70–0.98} {1.35–3.61}
2 {0.78–1.19} {0.99–3.77}
3 {0.77–0.95} {0.99–2.84}
two bridges. These results imply that the cross-frames in this skewed system, especially
with a contiguous layout, generally experience slightly higher stress ranges than the normal
and horizontally curved systems. Although not explicitly addressed in the table, it was also
observed in several instances that cross-frame stress ranges were dependent on its location
relative to the right lane (i.e., the typical drive lane for heavy truck traffic). The discussion
related to load-position effects and bridge geometry is expanded in Section 3.2 with respect to
the finite element studies.
3.2 Fatigue Loading Study

This section summarizes the analytical results related to the Fatigue Loading Study that was
previously introduced in Section 2.4. This study, which focuses on addressing Objectives (a)
and (b) of NCHRP Project 12-113, generally explores the fatigue loading model used for cross-
frame analysis and design. More specifically, the Fatigue Loading Study seeks quantitatively
based answers to five major questions that were outlined in Section 1.2. These questions, along
with the dedicated subsection herein, are as follows:
• What is the influence of composite bridge geometry (e.g., support skew, horizontal curvature)
and cross-frame layout (e.g., cross-frame spacing, staggered layout) on the load-induced
fatigue behavior of cross-frame systems (including governing force effects, critical lane load-
ing, and critical members under AASHTO fatigue loading criteria)? [Section 3.2.1]
• Based on the findings of the item above, is it necessary to perform a refined analysis, either
simplified 2D or 3D methods, for straight and non-skewed bridges? [Section 3.2.1]
• Is the current fatigue load model in terms of truck position (i.e., single design truck passages
positioned in various transverse lane positions) appropriate for cross-frame analysis and
design? Or do multiple presence effects need to be considered? [Section 3.2.3]
• Is the “n” value (i.e., number of cycles per truck passage) currently assumed for the generic
“transverse member” designation applicable for cross-frames? [Section 3.2.2]
• Are the current AASHTO LRFD Fatigue I and II load factors that were calibrated for girder force
effects and recent WIM data appropriate for cross-frame analysis and design? [Section 3.2.2]
As noted in the list above, these questions are systematically addressed throughout this
section of the report. In the context of AASHTO fatigue loading criteria, Section 3.2.1 inves-
tigates the correlation between key bridge parameters (e.g., support skew) and the load-induced
force effects in critical cross-frames. These results are further evaluated to determine if and
under what circumstances live load effects can be ignored in the design of cross-frames (i.e.,
cases in which live load force magnitudes are insignificant compared to other load cases).
Sections 3.2.2 through 3.2.3 focus on cross-frame response to measured WIM records.
Section 3.2.2 examines the cross-frame response to the WIM records, namely the number of
stress/force cycles induced by the various load spectra, and investigates the use of modified

fatigue load factors for cross-frame design. Section 3.2.3 reviews the WIM data and explores
the frequency at which a “double truck” case (i.e., a loading condition that maximizes the force
reversal in a cross-frame member) occurs. Section 3.2.4 provides a summary of the key findings.
3.2.1 Influence of Bridge Geometry

As noted above, this subsection addresses the influence of bridge geometry on the live load
demands in the cross-frame members in the context of the AASHTO fatigue loading criteria.
More specifically, four different aspects are summarized from the study. First, cross-frame force
effects are compared on a member-by-member basis in Section 3.2.1.1. That is, the research
team investigated which member (i.e., type and location on span) typically governs the load-
induced fatigue design of cross-frames. Next, the transverse lane passage that maximizes load-
induced cross-frame forces is examined in Section 3.2.1.2. Section 3.2.1.3 studies the impact of
skewed and/or curved geometries on those force effects, and Section 3.2.1.4 evaluates the fatigue
loading criteria with respect to the fatigue resistance criteria (i.e., the cross-frames are effectively
designed based on 9th Edition AASHTO LRFD guidance). By effectively designing the cross-
frames, the overall efficiency and inherent conservatism of the AASHTO Specifications in terms
of load-induced cross-frame fatigue design can be assessed.
3.2.1.1 Governing Cross-Frame Member

To examine which cross-frame member typically governs load-induced fatigue design, the
research team evaluated the unfactored design force computed for each selected cross-frame
member in the 4,104 unique bridges. From these data, the cross-frame members that most often
govern the design (i.e., the cross-frame with the largest force range recorded) were identified
based on the procedure outlined in Section 2.3.2.
This process is summarized in Figure 3-7. The process compiles the unfactored design force
ranges for all 68,352 cross-frames in the parametric study in the form of a box-and-whiskers
plot that indicates the maximum and minimum in the range as well as the median and first and
third quarter values. This is discussed in more detail below. Each box-and-whiskers component
is organized by cross-frame location and then further by cross-frame member. Thus, each figure
presents the results of 16 different types of cross-frame members. For the sake of clarity, cross-
frame identification is abbreviated. As far as cross-frame location, “ES” represents the cross-
frame panel near the end support (colored in red), “M-I” the interior bay near the maximum
positive dead load moment region (blue), “M-E” the edge bay near the maximum positive
dead load moment region (green), and “IS” near the intermediate support (purple). Refer to
Section 2.3.1.5 for more information on these general locations.
In terms of the cross-frame members, “TS” represents top strut members, “D1” for diagonal 1
(diagonal framing into the top flange of the left girder where left is based on the coordinate
system previously established), “D2” for diagonal 2, and “BS” for bottom struts. Note, that for
K-type cross-frames, there are two independent bottom strut members. For purposes of the
results hereafter, only the governing member is presented for consistency.
The plots are also organized by bridge type to illustrate the effects of support skew and bridge
curvature on the cross-frame response. For instance, data labeled as “straight, normal” was
generated from all bridges with no bridge curvature (i.e., infinite radius of curvature) and zero
support skew. To fully understand and interpret the box-and-whiskers plot, there are additional
items to consider:
• Each box-and-whisker segment represents the results of all bridges and cross-frames that
satisfy the applied filters (i.e., a probability design function). For example, the box-and-
whiskers labeled “TS” under the “M-E” category in the “Straight, Normal” graph represents

Figure 3-7. Unfactored design force ranges compiled for every cross-frame
member evaluated.
the compiled design force ranges for all top strut members evaluated near the maximum
positive dead load moment region (edge bay) of straight and normal bridges. In total,
312 data points (corresponding to the unfactored design force ranges for 312 unique, straight
and normal bridges) are represented.
• For each box-and-whiskers plot, six key statistical parameters are displayed: (i) the minimum
value in the data set, (ii) the first quartile or the 25th percentile, (iii) the second quartile or
the median value, (iv) the third quartile or the 75th percentile, (v) the maximum value, and
(vi) the mean value. The maximum and minimum values are represented by the whiskers, the
quartile values are represented by the bounds of the box, and the “X” represents the mean.
A graphical representation of this procedure is presented in Appendix F for reference.
• The critical lane passage that corresponds to the design force ranges presented are different
for each point in the data set. In this figure, only the magnitudes of the design force ranges are
of interest. The critical lane positions are covered in a subsequent section.
• A reference sketch is also provided in the figure to clarify the relative position of the cross-
frame considered; the results for the color-coded cross-frame locations are plotted. Note
that the exact location of the cross-frame, the number of girders, the number of spans, the
support skew angle, and the radius of curvature in the sketch are for visual reference only;
the bridge geometries of the various data points may differ from the representative sketch.
By presenting the box-and-whiskers side-by-side, the cross-frame members that generally
experience the largest force ranges due to the passage of the AASHTO fatigue truck in its critical

lane can be evaluated. By examining the results independently for bridge types, the impacts of
support skewness and bridge curvature can also be assessed.
From Figure 3-7, the following observations can be made about governing cross-frame
members:
• There is substantial variability observed in all the box-and-whiskers plots presented. For
example, the design forces for bottom struts near end supports of horizontally curved and
skewed bridges (curved, skewed) range from a minimum of about 0.5 kips to a maximum of
nearly 18 kips. This scatter is attributed to the large number of bridge parameters inherently
considered in the results.
• Top strut members, as anticipated, generally have the lowest force demands observed since
the concrete slab tends to restrain out-of-plane deformations at the top of the girder.
• The force ranges observed in skewed and/or horizontally curved bridges are generally greater
than equivalent straight and normal bridges, particularly in bottom strut and diagonal
members near skewed supports. For example, the mean force range for bottom struts at end
supports (BS; ES) increases from 3 kips for straight and normal bridges to over 4 kips for
straight and skewed bridges.
• This same relationship is not as impactful for cross-frames not in the vicinity of supports
(i.e., near the maximum positive dead load moment region), as cross-frames in these regions
are less impacted by the effects of support skew.
• In general, bottom strut members near M-I regions (interior bay near maximum positive
dead load moment region) have the largest design force ranges for non-skewed (straight or
horizontally curved) bridges. The maximum, 75th percentile, and mean values are all largest
for this condition.
• For skewed bridges, the bottom strut members near end supports tend to have the largest
force demands, but there is still significant scatter.
From all the data obtained from this study, it is apparent that bottom strut members tend to
be the most critical in terms of fatigue force demands. In past studies that focused on stability
bracing applications of noncomposite girders, it has been shown that the top and bottom struts
generally behave as zero-force members with both diagonal members equally effective (i.e.,
adjacent girders rotating equally). However, for cross-frames in composite systems and sub-
jected to live loads, bottom struts are often heavily engaged. In many cases, the cross-frames
behave similarly to floorbeams in a stringer-floorbeam system to distribute loads from girder-
to-girder, where adjacent girders rotate differentially in this case. The bottom strut is analo-
gous to the bottom tension flange of a composite floorbeam, the concrete deck is the top
compression flange, and the top strut is in close proximity to the pseudo-neutral axis.
At skewed supports, the global, torsional response of the superstructure tends to engage the
nearby bottom struts. Contiguous lines of cross-frame panels act like a stiff, closed section that
resists the torsional moments on the bridge cross-section.
Even with these discernible trends, there is still significant variability in cross-frame response.
The governing cross-frame member is still a function of many bridge parameters, albeit support
skew and curvature are the most important. Given a random bridge geometry, the critical
cross-frame panel could likely be identified within reasonable limits before any analysis is
performed. Still, it is recommended that the critical cross-frame is not “missed” by taking a
shortcut. Rather, conducting an influence-surface analysis and performing a comprehensive
design of all intermediate cross-frames in the bridge ensures a fully vetted design.
Many commercial design software packages have the built-in functionality to analyze and
design all cross-frames for design loads. If the program does not automatically address this,

manually developing a spreadsheet to evaluate each cross-frame due to AASHTO fatigue loading
criteria is possible.
3.2.1.2 Governing Lane Passage

Where Section 3.2.1.1 evaluated which cross-frame members govern load-induced fatigue
design, the section herein addresses the lane position corresponding to those critically loaded
members. As currently documented in AASHTO LRFD Article 3.6.1.4.3a, bridge designers
must consider all transverse and longitudinal truck positions when evaluating details and
components for the fatigue limit state. Other than a 2-foot clear distance from the inside face of
the bridge barrier, the AASHTO fatigue truck passages must be evaluated in all possible trans-
verse lanes, forward and backward—a provision that is included primarily due to the uncertainty
in future lane striping and/or bridge widening. Therefore, without assurance that the traffic
lanes are to remain unchanged throughout the service life a bridge, the designer must take the
conservative approach.
As documented previously, the AASHTO fatigue truck traversed over all 4,104 bridges
in 1-foot increments of transverse lane positions. Given the number of lane passages and
cross-frame members studied, drawing concise and meaningful conclusions on the data was
challenging. In the evaluation of the data, however, the research team observed that many
of the critical cross-frame members were governed by truck passages along the inside faces of
the barriers (i.e., centerline of the wheel line within 1 foot of the closest barrier). The minimum
distance of 1 foot from the inside faces of the barriers was conservatively used in this study as
described previously.
Thus, to simplify the discussions in this section, lane passages are grouped as either
“overhang” loads or “non-overhang” loads. Overhang loads represent truck passages where
one of the transverse wheel lines is applied outboard of the fascia girder centerline; these
load cases include “left” and “right” overhang loads, as well as both the forward and backward
directions. Non-overhang loads represent truck passages for which both wheel lines are within
the centerlines of the fascia girders.
Overhang loads, as noted above, occur on a much less frequent basis for bridges in service.
Although these loading cases are often critical for strength limit states, truck drivers are less
likely to frequently drive within a few feet of the barrier (i.e., along the overhanging portion
of the deck). As such, overhang loads are less representative of the fatigue limit state, which
focuses heavily on loading most likely to occur at a high frequency throughout the design life.
Non-overhang loads, on the other hand, are much more representative of the fatigue limit state.
With that said, the focus of this subsection is to identify the types of bridges most commonly
governed by overhang load cases. Assuming the designer is uncertain about future lane striping
and/or bridge widening, these special bridge types are the most affected by the relatively conser-
vative language in Article 3.6.1.4.3a.
Figure 3-8 compiles the lane position associated with the results presented in Figure 3-7.
For each of the 4,104 unique bridges, the critical lane passage corresponding to the governing
cross-frame member was recorded and was later categorized as an overhang or non-overhang
load based on the definition above. The tallied results are presented in the form of a bar graph
in Figure 3-8, where the y-axis represents the frequency of occurrence with respect to the
number of bridges in each type (e.g., 312 straight and normal bridges). Each of the bars is then
further delineated as overhang and non-overhang loading conditions.
The figure independently evaluates bridge types, similar to Figure 3-7, but does not explicitly
present data from individual member types (i.e., bottom struts, diagonals, top struts). Instead,
any of the governing cross-frame members from Figure 3-7, regardless of member type, are
grouped based on their general location in the framing plan (i.e., ES, M-I, M-E, and IS).

Reference Sketch:
ES M-I
IS Key:
Non-Overhang Load
M-E Overhang Load
1.0
312 Models 1,440 Models 624 Models 1,728 Models
0.9
0.8
Frequency of Occurrence
M-I
0.7
ES
0.6
M-I
M-E
0.5
IS
0.4
ES
M-E
0.3
M-I
M-E
0.2
ES
M-E
IS
M-I
0.1
IS
ES
IS
0.0
Straight, Normal Straight, Skewed Curved, Normal Curved, Skewed
Bridge Type
Figure 3-8. Bar graph presenting the critical lane position associated
with the governing cross-frame member.
To clarify the intent of the figure, the straight, normal data set is considered as an example.
For all 312 straight and normal bridges evaluated, the governing cross-frame was located in an
interior bay near the maximum positive dead load moment region on 204 occasions (65%). Of
those 204 occasions, the critical cross-frame member was always governed by a non-overhang
load (i.e., 100% non-overhang and 0% overhang). This is reflected in the “M-I” bar in the
straight, normal portion of Figure 3-8.
A more interesting example is the intermediate support “IS” case of the straight and skewed
data set. In total, 1,440 straight and skewed bridges were analyzed in the Fatigue Loading
Study. Of those 1,440 bridges, 565 were governed by cross-frame members near the intermediate
support (39%). Of those 565 cases, 514 were governed by overhang truck passages and only
51 by non-overhang loads.
From Figure 3-8, the following observations can be made about bridge type and critical
lane passages:
• Non-overhang loads, although not overly descriptive, generally correspond to “localized”
load effects. In other words, the force effects in the critical cross-frame were maximized by
a truck passing just to the left or right of the panel, similar to the trends observed in the
influence-surface plots in Section 2.3.
• For non-skewed bridges, the critical cross-frame members are almost always governed by
non-overhang loads. In fact, the forces in critical cross-frame members in straight, normal
bridges were maximized 100% of the time by non-overhang loads. The exception to this rule
is M-I cross-frames in horizontally curved bridges (interior bay near the maximum positive
dead load moment region), which were often governed by truck passages along the outer
radius of the curve. Thus, it can be concluded that cross-frame fatigue design in bridges
without support skews is largely governed by truck passages in close transverse proximity to
the cross-frame panel.
• For skewed bridges, the overhang loads represent a large percentage of the critical truck
passages, particularly for “ES” and “IS” cross-frames. The torsional response of the super-
structure and the corresponding girder rotations often heavily engage the bottoms struts

and diagonals in these areas. As such, loads that maximize the induced torque on the bridge
cross-section are often critical. For straight bridges, the critical loads correspond to those
applied along either deck edge where the moment arm about the bridge centroid is largest.
For horizontally curved bridges, those loads correspond to either the inner or outer radius
about the straight-line chord of the curved segment.
It is evident that force demands in cross-frames of skewed bridges, especially those near the
end or intermediate supports, are sensitive to overhang loads. Because these cases are rather
infrequent, basing a fatigue limit state design on an overhang truck passage could potentially
result in overly conservative design loads. With that in mind, overhang and non-overhang truck
passages are evaluated in the context of measured WIM data in Section 3.2.2, where the appro-
priate lane position for AASHTO implementation is examined in greater depth.
3.2.1.3 Impact of Bridge Parameters

As discussed in the preceding subsections, there is significant variability in load-induced
cross-frame response. This was particularly evident in Figure 3-7, which presented the governing
force demand in every cross-frame member evaluated in the 4,104-model matrix. The variability
in the response is attributed to many of the parameters that comprise the bridge structures.
This section herein examines the impacts of these parameters on the force effects in critical
cross-frame members.
Before evaluating the effects of each parameter, Figure 3-9 presents one isolated example
for reference. In this plot, the response of a diagonal cross-frame member near the maximum
positive dead load moment region is illustrated for three different bridge structures. The
influence line shown corresponds to a passing AASHTO fatigue truck, whose left wheel line is
positioned 5 feet from the bridge centerline. Therefore, the truck passes just to the left of the
cross-frame panel of interest, as depicted in the framing plan sketch in Figure 3-9. The three
bridges share identical parameters except for curvature: straight, 1,500-ft radius, and 750-ft
radius. Thus, any differences in the cross-frame response can be attributed directly to the
introduction of a curved layout.
Based on this sample data, it is evident that the force demand in the cross-frame member
of interest increases as the bridge radius decreases (or its curvature increases). Relative to the
D2
8
750-ft Radius Radius of Force Range
1500-ft Radius Curvature (ft) (kips)
6
750 7.00
Axial Force (kips)
7.0 kips 1,500 6.45

4 Straight 5.82
Straight
2
-2
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360
Figure 3-9. Sample data demonstrating the impacts of bridge curvature

on cross-frame force demands.

straight bridge, the resulting force range from the passing truck increases 10% when a 1,500-ft
radius is introduced and an additional 9% when the radius is halved to 750 feet. The general
response of the cross-frame is the same, except for the magnitudes. Thus, based on this limited
data set, it could be stated that bridge curvature is positively correlated to cross-frame force
demands (i.e., bridge radius is negatively correlated).
Figure 3-9 represents only a select number of unique conditions. The response could also
have been presented for different lane positions, different cross-frame members, and differ-
ent bridges altogether. Ultimately, a primary interest of most designers is knowledge of how a
bridge parameter impacts the overall fatigue design of cross-frames, not the localized effects on
non-governing members that are deemed less important. Therefore, the results presented here-
after are focused on the governing cross-frame members and lane positions (i.e., the unfactored
design forces in accordance with current AASHTO fatigue loading criteria).
To expand on the sample results presented in Figure 3-9, critical cross-frame force effects
and lane passages were evaluated for every bridge in the analytical testing matrix. That is to say,
box-and-whiskers plots that represent the spectrum of design forces were generated and sub
sequently categorized by bridge parameter (e.g., all cross-frame force effects in straight bridges,
bridges with a 1,500-foot radius of curvature, and bridges with a 750-foot radius of curvature
were evaluated and compared).
Figure 3-10 presents the summarized results from this exercise highlighting several key
parameters, including skew index (defined below), horizontal curvature, cross-frame layout,
number of girders, and concrete deck thickness. These parameters were selected, as they signi
ficantly influence the load-induced force effects in the critical cross-frames. In these figures,
box-and-whiskers plots and response spectra are represented as a series of three distinct line
graphs. In the figure, the solid black line represents the mean of each data set, and the dashed
lines represent the 95th and 5th percentile values. The respective percentile lines give an
indication where 90% of the data set is populated. Despite only presenting five parameters
in Figure 3-10, every independent variable that was previously identified in Table 2-2 was
evaluated similarly. For additional information on the development of these plots and the
results for all bridge parameters (including spot checks of bridge models without barriers),
refer to Appendix F.
Before introducing the results, it is prudent to define skew index, as well as connectivity
index. Skew (Is) and connectivity (Ic) indexes are respective measures of bridge skewness and
curvature, as defined in the AASHTO G13.1 Guidelines for Steel Bridge Analysis (2019). These
indexes are used to categorize bridge geometries for purposes of recommended analysis practices.
Bridges with larger skew and connectivity indexes require more advanced analysis procedures,
either improved 2D or 3D techniques. Although not presented in Figure 3-10, the connectivity
index is utilized in subsequent sections of the report. The skew and connectivity indexes are
defined by the following expressions:
wg tanq
Is = 3.1
Ls
15,000
Ic = 3.2
R (ncf + 1)m
where
Is, Ic = skew index and connectivity index, respectively,
wg = width of bridge, measured between the centerlines of the fascia girders,

18 18
Skew Index Curvature
15 15
12 12
9 9
6 6
3 3
0 0
0 0.35 0.7 0 5 10 15
Skew Index Curvature (10–4/ft)
18 Cross-Frame Layout 18 No. of Girders

15 15
Max Force Range (kips)
12 12
9 9
6 6
3 3
0 0
3 5 7
Cross-Frame Layout No. of Girders
18 Deck Thickness
15
12
0
8 10
Deck Thickness (in)
Figure 3-10. Unfactored design force spectra demonstrating the impacts

of several key bridge parameters.
q = largest skew angle on bridge relative to the axis normal to the longitudinal girders,
Ls = span length at the bridge centerline,
R = minimum radius of the horizontal curvature,
ncf = number of intermediate cross-frames in the span, and
m = constant, taken as 1 for simple-span bridges and 2 for continuous-span bridges.
As the equations show, the skew index increases for shorter, wider spans with larger skew
angles. Similarly, the connectivity index increases for tighter curves (i.e., smaller radius of
curvature) and fewer cross-frames connecting the girders together.
From Figure 3-10, the following observations can be made about cross-frame force demands
and the select bridge parameters outlined above:
• Skew index: It is evident from the results that skew index substantially impacts the response of
cross-frames. A +0.52 correlation coefficient was obtained, indicating a strong, positive
relationship between skew index and cross-frame force effects. In simpler terms, a +0.1 increase

in skew index generally resulted in a 25% increase in the critical cross-frame force demand.
In general, these observations are consistent with the field experimental data outlined in
Section 3.1.
• Curvature: A positive correlation is observed between bridge curvature and cross-frame force
effects (i.e., correlation coefficient of +0.32). On average, an increase in curvature of 10-4/ft
results in an increase of 3% for the governing cross-frame force effect. Thus, the trend observed
in Figure 3-9 for a small sample set is consistent with the full data set.
• Cross-frame layout: Staggering the cross-frame layout in skewed bridges significantly reduces
the governing design force in cross-frames (35% on average). This trend is attributed to the
stiffness reduction associated with a discontinuous line of cross-frames. These results are
consistent with the guidance provided in AASHTO LRFD Article C6.7.4.2.
• Number of girders: A positive correlation coefficient of +0.60 was observed between number
of girders (i.e., bridge width) and cross-frame force demands. A more variable response
was also observed as the number of girders increased, as evidenced by the distance between
the 95th and 5th percentile lines. On average, increasing the overall width of the bridge
results in a 30% increase in the governing force effect. These trends are attributed to the
overhang truck loads outlined in Section 3.2.1.2. Eccentric loading and torque on a straight
or curved system are amplified for wider bridges, which in turn amplifies the governing
cross-frame response.
• Deck thickness: Deck thickness and cross-frame force effects are negatively correlated
(–0.22 correlation coefficient). On average, a 1-inch increase in deck thickness results in a
14% decrease in force demands. A reduction in deck thickness often indicates that the
cross-frames will attract a higher proportion of the load distribution, which increases member
force effects. Additionally, a thinner deck corresponds to a reduction in overall superstructure
stiffness, which subsequently increases girder displacements and cross-frame deformations.
In general terms, it is evident that skewed and/or curved composite bridge systems typically
result in larger load-induced force demands in their cross-frames. Although this behavior
has been widely recognized by designers throughout the last few decades, the results above
validate these assertions quantitatively. To mitigate these force effects and potential load-
induced fatigue cracking problems in new designs, designers can utilize these trends to avoid
an iterative “chase-your-tail” design solution (i.e., increase the size of the cross-frame member to
accommodate the design forces, re-analyze, and redesign for larger forces in the second pass, etc.).
For instance, using a discontinuous, staggered cross-frame layout is a practical and economical
solution for skewed systems. Increasing the deck thickness or decreasing girder spacing can also
lessen load-induced cross-frame forces in composite bridges, but adjusting those parameters
has significant impacts on the rest of the design.
3.2.1.4 Design Parameters

Based on the procedures outlined in Section 2.3.2, governing cross-frame force effects were
subsequently factored and converted to axial stresses, and AASHTO fatigue resistances were
evaluated. By effectively designing the cross-frames in accordance with current AASHTO LRFD
Specifications, the overall efficiency and inherent conservatism of the specifications in terms
of load-induced cross-frame design can be assessed.
Figure 3-11 presents the factored Fatigue II stress ranges that govern the design of all
4,104 representative bridges in the form of a histogram. The stress ranges were categorized in
0.1-ksi bins, and the counts were summed. For reference, the factored design stresses for the
three instrumented bridges (based on analytical procedures outlined previously) as well as the
measured effective stress ranges (Section 3.1.2) are also included. Recall that the Fatigue II load
effects derived from AASHTO LRFD are synonymous with the effective stress range of the truck
population; as such, a direct comparison can be made between the two entities. By comparing

200
Bridge 1 Key:
Bridge 3 Fatigue II “Analytical” Design Stress
160
Number of Occurrence
Bridge 2 Bridge 2 Measured Effective Stress
120 Bridge 1
Bridge 3
80
40
0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0
Governing Fatigue II Design Stress, γΔf (ksi)
Figure 3-11. Governing cross-frame fatigue stress ranges for

all bridges in Fatigue Loading Study.
analytical and experimental results of instrumented Bridges 1, 2, and 3, a few key observations
are established.
First, the governing design stress ranges for Bridges 1 and 2 are close to the mean response
(2.35 ksi), and the stress range for Bridge 3 is in the lower tail of the spectrum. Second and
more importantly, it is evident that the analytical results generally exceed the measured results
by a considerable margin. This is a preliminary indication that the AASHTO fatigue model for
the Fatigue II limit state may be conservative when compared to real loading conditions expe-
rienced by the instrumented bridges in the context of cross-frame behavior. This assertion is
verified for a wider range of bridges herein; however, note that the physical evidence from
measured data, rather than just relying entirely on analytical data, strengthens this observation.
Figure 3-12 then illustrates the governing demand-to-capacity ratio (i.e., DFn/γDf) with
respect to the Fatigue II limit state. The factored stress ranges from Figure 3-11 are compared
to the factored resistances computed based on the procedure outlined in Section 2.3.2. Demand-
to-capacity (D/C) ratios below unity indicate designs in conformance with AASHTO LRFD,
0.1
Llano, TX, Traffic
0.08 Mean = 0.54
% Above 1.0 = 7.4%
0.06
0.04
Probability of Occurrence
0.02
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
0.1
Houston, TX, Traffic
0.08 Mean = 1.40
% Above 1.0 = 65%
0.06
0.04
0.02
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Demand-to-Capacity Ratio
Figure 3-12. Fatigue II cross-frame demand-to-capacity ratios

for two extreme traffic conditions.

whereas ratios exceeding unity represent designs in violation of AASHTO LRFD. As outlined
previously, it is assumed that the design of all intermediate cross-frame members is governed
by the critical, maximum case. Thus, any bridge exceeding a D/C of 1.0 means that one
(i.e., the critical cross-frame member) or more members are inadequate in terms of Fatigue II
design criteria.
Figure 3-12 presents D/C histograms for the two extreme traffic conditions: rural Ranch
Road 152 in Llano, Texas, and urban I-10 in Houston, Texas. As outlined in Section 2.3.2, in these
examples, the Llano and Houston bridges are just hypothetical locations of a rural bridge with
relatively light traffic versus a busy urban environment with relatively high traffic. Based
upon TxDOT data (2020), the single-lane ADTTSL for the Llano location can be estimated as
182 compared to 3,000 for the Houston location. For each individual histogram, the factored
demands are identical; however, the factored capacities differ based on the large disparities
in projected truck traffic volumes.
From Figure 3-12, it is apparent that the projected traffic conditions have a significant
impact on the results. For the low-volume conditions, the results indicate that the majority
of the 4,104 bridges are conservative in terms of cross-frame fatigue design. In fact, the average
D/C ratio was 0.54, which implies substantial reserve capacity for the majority of bridges.
Only 7.4% of the models exceeded a D/C ratio of unity.
In contrast, the results indicate a more severe trend for the high-volume conditions under
the Fatigue II limit state (i.e., the 75-year ADTTSL does not exceed 8,485 trucks per day for a
Category E′ detail per AASHTO LRFD Table 6.6.1.2.3-2). On average, the D/C ratios exceeded
1.0 (mean of 1.40), and 65% of the bridges resulted in a design in violation of AASHTO
LRFD. If these bridges were to be implemented for a real construction project, modifications
to the cross-frame properties (e.g., an increase in cross-sectional area) or layout (e.g., use a dis-
continuous cross-frame layout in heavily skewed bridges) would likely be required to bring the
design in conformance with AASHTO LRFD. To expand further on Figure 3-12, Figure 3-13
highlights the differences between straight and curved bridges, as well as bridges with normal
and skewed supports. The full histogram related to the Houston, Texas, traffic in Figure 3-12 is
broken down into four bridge types: straight bridge with normal supports, straight bridge with
skewed supports, curved bridge with normal supports, and curved bridge with skewed supports.
Each subset histogram (shown in black) is overlaid on the full histogram (shown in light blue)
to demonstrate how bridge curvature and support skewness affect the D/C ratios.
The average D/C ratio increases when skewed supports and/or horizontal curvature are
introduced. For instance, the mean D/C for straight, normal bridges is 0.88; that value increases
to 1.23 for straight, skewed bridges and 1.15 for curved, normal bridges. A similar trend is
observed for the percent unconservative metric.
In broader terms, the results from Figure 3-12 and Figure 3-13 suggest the following:
• For roadways with low truck traffic, load-induced fatigue does not appear to be a major
concern for the 4,104-model bridge sample (which represents the most common bridge
conditions in the United States) according to AASHTO 9th Edition criteria.
• For bridges serving heavy truck traffic, current AASHTO criteria indicate a potential load-
induced fatigue problem.
Development of appropriate design criteria for cross-frames, like other structural elements,
requires achieving acceptable structural safety but must also consider economy. As learned from
the industry survey discussed in Section 2.1 and Appendix D, actual load-induced fatigue prob-
lems in cross-frame members have seldom been documented by bridge owners.
Thus, experience has shown that cross-frames are largely satisfying structural safety
requirements in terms of load-induced fatigue. A question of interest then is whether further

0.04
Straight, Normal Straight, Skewed
Mean = 0.88 Mean = 1.23
0.03 % Above 1.0 = 35% % Above 1.0 = 54%
0.02
0.01
0
0.04
Curved, Normal Curved, Skewed
Mean = 1.15 Mean = 1.73
0.03 % Above 1.0 = 57% % Above 1.0 = 83%
0.02
0.01
0
0 1 2 3 4 5 0 1 2 3 4 5
Figure 3-13. Fatigue II cross-frame demand-to-capacity ratios

for various bridge types (Houston I-10 traffic).
economies are possible in cross-frame design, while still providing adequate structural safety.
The absence of observed load-induced fatigue failures does not necessarily imply the current
AASHTO LRFD design criteria are overly conservative and wasteful from a cost perspective.
Nonetheless, with a goal of developing improved design criteria for cross-frames, the question
of whether current AASHTO LRFD design criteria for load-induced fatigue of cross-frames
are too conservative is important to examine. The analysis presented earlier in this section
provides at least some indication that AASHTO LRFD may be too conservative and merits
closer scrutiny.
When considering whether AASHTO LRFD may be too conservative for load-induced
fatigue design of cross-frames, the following three factors may be potentially considered:
1. The AASHTO fatigue resistance model is potentially too conservative (DFn is too low). The
resistance model primarily consists of detail Category E′ and its associated constants
(Table 6.6.1.2.3-1).
2. 3D FEA models perhaps consistently produce overly conservative force predictions. Among
many modeling assumptions, the primary focus is the stiffness modification factors
(R-factors) for cross-frames in 3D models (Article C4.6.3.3.4).
3. The AASHTO fatigue loading model is potentially too conservative (γDf is too high). The load-
ing model comprehensively consists of the following: fatigue truck (Article 3.6.1.4.1), truck
positioning (Article 3.6.1.4.3a; Article C6.6.1.2.1), dynamic load allowance (Article 3.6.2),
load factors (Article 3.4.1), and shear lag factor [Table 6.6.1.2.3-1 (Condition 7.2)].
Note that, by simply listing the above factors, it is not advocated that changes to the design
specifications are necessary related to all these factors. The list is intended to be comprehensive
in nature and identify all possible sources. For instance, this research does not suggest that
welded cross-frame connections be reclassified as a higher fatigue category.

As noted previously, fatigue resistance is beyond the scope of the project and is not discussed
further. Instead, the research team narrowed its efforts to examining the modeling approach
(stiffness modification and simplified analysis techniques) and the fatigue loading model herein.
Section 3.3 studies the applicability of the current R-factor approach (0.65AE per AASHTO
Article C4.6.3.3.4) for composite systems. Section 3.4 explores the limitations of simplified 2D
modeling approaches. Sections 3.2.2 through 3.2.3 examine the loading model (primarily the
fatigue truck, truck positioning, and load factors) with respect to high-resolution WIM data.
3.2.2 Cross-Frame-Specific Load Factors

3.2.2.1 Fatigue I Stress Ranges (Normalized to Fatigue Truck)
As noted in Section 2.3.3, the governing Fatigue I stress range calculated for all WIM sites
was determined based on 99.99th percentile criteria. This stress range corresponds to the lowest
magnitude of the top 0.01% of all stress ranges recorded (for the governing cross-frame for
each bridge given the defined lane position). Figure 3-14 illustrates the selection of the Fatigue I
stress range from a CDF of stress ranges for a sample cross-frame member. In this figure, the
stress range corresponding to the largest stress range recorded is approximately 7 ksi; however,
the lowest stress range within the top 0.01 percent stress ranges is shown to be 5.15 ksi. Thus,
5.15 ksi is considered the maximum stress for which AASHTO Fatigue I load criteria is evaluated.
Once all Fatigue I stress ranges were calculated for each WIM record and each bridge (i.e.,
one value per bridge per WIM record), the stress ranges were normalized to the largest stress
range produced by applying the unfactored AASHTO design load (i.e., the fatigue truck) to
the same bridge in all possible transverse positions within the appropriate clear distance of the
barriers. Note that the positioning of the fatigue truck to create the largest stress range does
not necessarily correspond to the location of the WIM traffic stream. This is consistent with
the current design approach, where the bridge (and its cross-frames) is designed based on the
maximum force effect produced by locating the fatigue truck in the critical position between
the barriers (in accordance with AASHTO Article 3.6.1.4.3a).
The Fatigue I stress ranges and biases were calculated for each of the 20 bridges using all
of the high-resolution WIM site records, as well as the arithmetic mean of all stress ranges
produced by the 18 WIM site records and the “mean plus 1.5 standard deviations” of all stress
ranges produced by the 18 WIM site records (the standard deviation was calculated for the
1.0
0.9
0.8
Cumulative Design Function
0.7
0.6 0.99995
0.5 0.99990
0.4
0.99985
0.3
0.2 0.99980
0.1
0
0 1 2 3 4 5 6 7
Stress Range (ksi)
Figure 3-14. Illustration showing selection of 99.99th percentile

stress range.

distribution of all stress ranges produced by the 18 WIM site records). Load bias refers to the
ratio between the actual load effects and the design load effects (i.e., the WIM load effects
divided by the load effects caused by the fatigue design truck).
As documented in the SHRP 2 R19B report (Modjeski and Masters 2015), the value
associated with the “mean plus 1.5 standard deviations” was taken as both the bias of the data
(SHRP 2 R19B states this was a conservative measure, since it was unknown how accurately the
WIM records used in the project reflected truck traffic across the nation) as well as an appro
priate load factor. This resulted in a proposed load factor of 2.0 for Fatigue I; however, the load
factor adopted in the 8th Edition of AASHTO LRFD was 1.75. Based on discussions with indi-
viduals knowledgeable on the selection of this load factor, it appears that the value of 1.75 was
ultimately selected by using the arithmetic mean of the normalized stress ranges produced
by the WIM site records (i.e., the bias and load factor are both taken to be equal to the mean
bias value). This was rationalized by acknowledging perceived conservatism inherent in the
resistance data for Fatigue I and the fact that the WIM data reviewed was deemed to be
sufficiently abundant to represent national traffic loads.
Figure 3-15 summarizes the ratios of the 99.99th percentile WIM cross-frame stress ranges
for each of the 20 bridges divided by the unfactored fatigue truck stress range for each bridge.
Each bin value in this figure represents the average of all Fatigue I stress ranges for all WIM sites
applied to the identified bridge (as explained in the preceding paragraph). The solid line in this
figure represents the arithmetic mean of all Fatigue I stress ranges for all WIM sites applied to
the identified bridge. The dashed line in this figure represents the mean value of all “mean plus
1.5 standard deviations” (i.e., the mean plus 1.5 standard deviation for all values represented by
the individual bars).
Consistent with the approach that led to the 8th Edition AASHTO LRFD updates to the
Fatigue I load factor, the arithmetic mean value of all Fatigue I stress ranges for all WIM sites
would correspond to an appropriate Fatigue I load factor for cross-frames. In other words,
the stress ranges produced by the unfactored fatigue truck must be amplified by this factor
to produce an equivalent 99.99th percentile stress range that represents the 99.99th percentile
load effects caused by the WIM traffic streams. Without consideration of material fatigue resis-
tances (and therefore the reliability index), the mean values associated with the solid lines in
Figure 3-15 imply an appropriate load factor for the Fatigue I limit state is 1.01. The coefficient
1.75 Key:
Mean
1.50 Mean + 1.5 Standard Deviations
Maximum Normalized SR99.99%
1.25
(WIM/DesignTruck)
1.00
0.75
0.50
0.25
Representative Bridge Model ID
Figure 3-15. Summary of Fatigue I normalized 99.99% cross-frame

stress ranges for all WIM records/sites.

of variation associated with this value is 0.23. This load factor is less than the current Fatigue I
load factor of 1.75, which indicates a potential source of conservatism in the design load criteria
for cross-frame fatigue.
Recall from Section 2.3.3.1 that the GVWs for the WIM records are not normal; rather, many
WIM sites appear to have multi-modal distributions that likely correspond to natural groupings
of different vehicle types and payload. Appendix F discusses examining the distribution on a
normal probability plot in order to assess the normality of the data and applying a best fit line
to the CDF to estimate the mean and standard deviations. This procedure is performed for all
distribution summaries to compare directly to the means and standard deviations computed
using standard arithmetic formulas. Table 3-5 compares the standard deviations and means
obtained by both methods.
3.2.2.2 Fatigue II Damage Ratios (Normalized to Fatigue Truck)

As discussed in Section 2.3.3.3, the Fatigue II design criteria can be evaluated in terms of
accumulated damage, as opposed to simply the effective stress range of the truck popula-
tion spectra. The accumulated damage metric inherently considers both variable stress range
magnitudes and number of cycles. Thus, the total damage accumulated by the various WIM
streams on the critical, governing cross-frame members is compared directly to the damage
accumulated by the AASHTO fatigue truck (based on AASHTO criteria for the relationship
between number of trucks and number of cycles). Appendix F derives the expression used
for describing the fatigue damage ratio, l.Values of l less than 1 indicate that the damage
accumulated by the WIM data is less than that of the assumed values based on AASHTO
design criteria (i.e., the design criteria is overly conservative); the opposite is true to values
of l larger than 1. For the number of cycles used with the AASHTO fatigue truck in design,
the current AASHTO LRFD (Table 6.6.1.2.5-2) specifies a value of n equal to 1 for transverse
members spaced greater than 20 feet, and a value of n equal to 2 for transverse members
spaced equal to or less than 20 feet. For this project, the number of design cycles per truck
was taken as 1.0 regardless of the cross-frame spacing, which is consistent with the analytical
and experimental results obtained as part of this study.
Figure 3-16 summarizes the fatigue damage ratios of the WIM traffic streams for each repre
sentative bridge. Figure 3-16 is the same as Figure 3-15 with respect to the definition of the
dashed and solid horizontal lines.
Without consideration of material resistances (and therefore the reliability index), the value
associated with the “mean plus 1.5 standard deviations” implies that an appropriate load
factor for the Fatigue II limit state is 0.49. The coefficient of variation associated with this
value is 0.16. This load factor is significantly less than the 0.80 load factor currently specified
in AASHTO LRFD for Fatigue II. The use of the “mean plus 1.5 standard deviations” as an
indicator of an appropriate load factor is consistent with research documented in SHRP 2
Project R19B (Modjeski and Masters 2015) and the subsequent development of the 0.8 load
Table 3-5. Comparison of statistical parameters obtained for the Fatigue I

cross-frame stress ranges by two methods.
Coefficient Mean Plus 1 Mean Plus 1.5

Mean
of Variation Standard Deviation Standard Deviations
Value Obtained by
1.01 0.23 1.24 1.35
Direct Calculation
Value Obtained from
1.04 0.17 1.22 1.31
Normal Probability Plot

0.8 Key:
Mean
0.7
Maximum Fatigue DamageRatios

Mean + 1.5 Standard Deviations
0.6
(WIM/DesignTruck)
0.5
0.4
0.3
0.2
0.1
Representative Bridge Model ID
Figure 3-16. Summary of Fatigue II cross-frame damage ratios

for all WIM records/sites.
factor for Fatigue II. This suggests that the 0.8 load factor may be conservative for characterizing
the response of cross-frames to the U.S. truck spectra.
Similar to the comparison made in Section 3.2.2.1, the fatigue damage ratios for all WIM records
and all bridges were plotted on a normal probability plot to assess the normality of the damage
ratios (further discussed in Appendix F). Table 3-6 compares the standard deviations and
means obtained by standard arithmetic formulas and the normal probability plot approach.
3.2.2.3 Stochastic Simulation via the Monte Carlo Technique

A Monte Carlo simulation was performed using the statistical summaries for resistance and
cross-frame force effects previously described. For the simulation, a total of 10,000 samples
were randomly generated from the distributions of load and resistance described by the statis
tical parameters. The procedure used in the development of this simulation is discussed in
Appendix F and was performed for the cross-frame statistics in Table 3-7 for each detail
category. The resulting reliability indices are shown in Table 3-8.
Based on the reliability analysis performed, the resulting reliability indices for the Fatigue II
limit state for cross-frames using the reduced load factors introduced in Section 3.2.2.2 generally
exceed the target reliability assumed inherent within the AASHTO Specifications (b = 1), as
presented in the SHRP 2 R19B study (Modjeski and Masters 2015). In contrast, the result-
ing reliability indices associated with the reduced Fatigue I load factors introduced in Sec-
tion 3.2.2.1 vary, for which the indices of several detail categories are less than unity.
Table 3-6. Comparison of statistical parameters obtained for the Fatigue II

cross-frame damage ratios by two methods.
Coefficient Mean Plus 1 Mean Plus 1.5

Mean
of Variation Standard Deviation Standard Deviations
Value Obtained by
0.40 0.16 0.46 0.49
Direct Calculation
Value Obtained from
0.40 0.21 0.48 0.53
Normal Probability Plot

Table 3-7. Statistical parameters describing the load effects

in cross-frames for Fatigue I and II limit states.
Coefficient of
Limit State Bias for Load Data Load Factor
Variation of Load
Fatigue I 0.26 1.01 1.01
Fatigue II 0.21 0.40 0.53
In order to obtain reliability indices close to the assumed target reliability of unity, adjust-
ments are required to either the load or resistance factors. As an alternative to altering the
resistance factors [or the associated changes in constant amplitude fatigue thresholds, (F)TH,
for Fatigue I, or detail constants, A, for Fatigue II] it is also possible to adjust the load factor
for fatigue such that the minimum target reliability of unity is achieved for each category.
Additionally, as an alternative to introducing two new load factors for the fatigue limit state
for cross-frames, it is possible to apply a single adjustment factor to the existing load factors,
similar to the procedure used for fatigue design of orthotropic decks established in AASHTO
LRFD (2020) Article 3.4.4.
Using an adjustment factor of 0.65 applied to the Fatigue I load factor of 1.75 (i.e., a resul-
tant load factor for cross-frames of 1.14 for Fatigue I) and the Fatigue II load factor of 0.8 (i.e.,
a resultant load factor for cross-frames of 0.52), the resulting reliability indices are calculated
via Monte Carlo simulation and shown in Table 3-9. With this adjustment, each detail category
satisfies the minimum assumed target reliability of unity.
3.2.2.4 Reevaluation of AASHTO Fatigue Design Criteria

Based on the fact that reduced cross-frame-specific load factors have been proposed to
eliminate a source of conservatism observed in the fatigue loading model, it is worthwhile
revisiting Figure 3-13. Recall that Figure 3-13 evaluated the fatigue design performance of
cross-frames in each of the 4,104 bridges studied. Based on those results, it was apparent
that current AASHTO design criteria indicate a potential load-induced fatigue problem in
common cross-frame configurations, despite the lack of physical evidence in constructed super-
structures across the United States. As such, Figure 3-13 is replicated in Figure 3-17 with two
notable exceptions.
Table 3-8. Reliability of cross-frames for Fatigue I and II limit states using
the load factors introduced in Sections 3.2.2.1 and 3.2.2.2.
Reliability Index
Detail Category
Fatigue I Fatigue II
A 1.01 1.77
B 0.88 1.74
B′ 0.98 2.47
C 0.93 1.82
C′ 0.93 1.82
D 1.27 2.89
E 0.60 1.95
E′ 1.41 2.31

Table 3-9. Reliability of cross-frames for Fatigue I and II limit states using
adjustment factor of 0.65 to existing load factors.
Reliability Index
Detail Category
Fatigue I Fatigue II
A 1.32 1.74
B 1.22 1.70
B′ 1.49 2.39
C 1.29 1.79
C′ 1.30 1.77
D 1.82 2.81
E 1.08 1.87
E′ 1.77 2.25
0.08
Straight, Normal Slightly Curved
IS = 0 IS = 0
IC = 0 IC ≤ 1
0.06
Mean = 0.55 Mean = 0.62
% Above 1.0 = 5% % Above 1.0 = 9%
0.04
0.02
0.08
Heavily Curved Slightly Skewed
IS = 0 IS ≤ 0.3
IC > 1 IC = 0
0.06
Mean = 0.78 Mean = 0.66
% Above 1.0 = 19% % Above 1.0 = 16%
0.04
0.02
0.08
Heavily Skewed Skewed & Curved
IS > 0.3 IS > 0.1
IC = 0 IC > 0.5
0.06
Mean = 1.02 Mean = 1.19
% Above 1.0 = 44% % Above 1.0 = 56%
0.04
0.02
0
0 1 2 3 4 5 0 1 2 3 4 5
Figure 3-17. Fatigue II cross-frame demand-to-capacity ratios for various

bridge types (Houston I-10 traffic) using reduced load factors.

First, the Fatigue II load factors inherently built into the force demands (i.e., the numerator
of the D/C ratio) are reduced from 0.80 to 0.52 based on the findings outlined above. Note that
the resistance model (i.e., the denominator in the D/C ratio) remains unchanged. The second
major exception is in how skewed and curved bridges are organized and presented. Rather
than simply grouping the bridges based on the presence of support skew and/or horizontal
curvature, the indexes developed in NCHRP Research Report 725 (White et al. 2012) and
previously defined in Eqs. 3.1 and 3.2 are utilized.
Thus, each bridge in the 4,104-model data set is grouped based on the skew and connectivity
index bounds established in NCHRP Research Report 725. For instance, heavily curved bridges
are differentiated from moderately curved bridges by Ic > 1 and Ic ≤ 1, respectively. By orga-
nizing the figure in this manner, the criticality of a particular bridge type in terms of load-
induced fatigue forces in cross-frames can be clearly delineated. It is also important to note
that Figure 3-17 only investigates fatigue performance in the 4,104 bridges with respect to the
high-volume truck traffic in Houston, Texas. The results for low-volume Llano traffic would
be much less severe.
With consideration of the proposed cross-frame-specific load factors and I-10 Houston
traffic, the following observations can be made from Figure 3-17:
• In straight bridges with normal supports, the average D/C ratio was 0.55, where only 5% of
the 312 qualifying bridges exceeded unity. This indicates that load-induced fatigue would not
govern the design of cross-frames in these bridge types even for heavy truck traffic volumes.
This implies that conducting a 2D or 3D refined analysis to obtain live load force effects
in cross-frames is likely not warranted. This observation is consistent with the current
AASHTO LRFD design approach.
• For slightly curved bridges (Ic ≤ 1) the mean D/C ratio (0.62) increased slightly compared
to straight bridges, and only 9% of the qualifying data points exceeded unity. This indicates
that design of cross-frames in moderately curved bridges with normal supports is also not likely
to be governed by load-induced fatigue. In contrast, heavily curved bridges (Ic > 1) showed
an increase in mean D/C ratio to 0.78 and an increase in percent exceeding unity to 19%.
• It is evident that support skew affects cross-frame force demands more significantly than
horizontal curvature. Slightly skewed bridges (Is ≤ 0.3) and heavily skewed bridges (Is > 0.3)
produced mean D/C ratios of 0.66 and 1.02, respectively. Additional statistical analysis
showed that obtaining fatigue loads and designing for load-induced fatigue should be
considered for cross-frames in bridges with a skew index exceeding 0.15.
• For bridges with both support skew and horizontal curvature (i.e., Ic > 0.5 and Is > 0.1), the
mean D/C reported was 1.19, which indicates a significant load-induced fatigue issue even
with the reduced load factors. Consequently, these bridges would likely need to be redesigned
or the cross-frames reconfigured to mitigate the fatigue force demands. A refined 2D and
3D analysis would be necessary to accurately obtain those design forces.
3.2.3 Multiple Presence Study

The assumptions of multiple presence in the original LRFD calibration studies (Nowak 1999,
Kulicki et al. 2007) were initially based on engineering judgement and visual observations
of truck traffic with unknown weights. These initial assumptions were that a “side-by-side”
scenario (i.e., adjacent lane loaded with a passing truck in general alignment with the drive
lane truck) occurred once every 15 load events. More recent studies have shown this assump-
tion to be excessively conservative (Sivakumar, Ghosn, and Moses 2011). While the research
conducted in SHRP 2 R19B (Modjeski and Masters 2015) confirmed this is an unlikely event,
prior to this study the effect of multiple presence has not been reviewed in the context of
cross-frame force effects.

450 Indiana US 31 SB 120%
400
100%
350
300 80%
Key:
250 Frequency
Cumulative % 60%
200
150 40%
100
20%
50
0 0%
-
-
-
-
-
-
-
-
-
Clear Distance Between Trucks (ft)
-
Figure 3-18. Histogram from two-lane WIM records showing

the clear distance between a passing truck and the drive lane
truck (Indiana US-31 SB).
As discussed in Section 2.3.3.5, a cluster analysis was performed on the multi-lane WIM data
to consider if a bridge may be loaded with other truck traffic during a primary drive lane load
event (i.e., passage of one or more axles of a vehicle). Typical results of the adjacent lane
scenarios are summarized in Figure 3-18 through Figure 3-20, which provide a sample histo
gram illustrating various aspects of multiple presence studied. For various load-position
parameters (e.g., clear distance between drive lane and passing truck), the number of occur-
rences for each multi-lane WIM site is compiled and plotted. Specifically, Figure 3-18 illustrates
the clear distances between passing lane trucks and drive lane trucks, Figure 3-19 illustrates
the spectrum of GVW for passing lane trucks, and Figure 3-20 illustrates the ratio of passing
lane truck GVW to drive lane truck GVW. Appendix F provides additional histograms from
these studies. Note that the occurrences on the primary vertical axis for these histograms are
the number of total occurrences for the 12-month data set. This value should be taken into
context with the total annual truck traffic (ATT) for the specific site. Table 3-10 summarizes
2500 Indiana US 31 SB 120%
100%
2000
Key: 80%
1500 Frequency
Cumulative % 60%
1000
40%
500
20%
0 0%
Passing Lane GVW (kips)

GVW of a passing truck (Indiana US-31 SB).

1600 120%
Indiana US 31 SB
1400
100%
1200
Key: 80%
1000
Frequency
Cumulative % 60%
800
600
40%
400
20%
200
0 0%
Ratio of Passing Lane GVW to Drive Lane GVW

the ratio of the passing lane GVW to the drive lane GVW
(Indiana US-31 SB).
how often (for the year of data considered) any truck was within a certain window of the
drive lane truck, relative to the total volume of traffic in the drive lane for each specific site.
Based on these results, it is evident that a passing truck in close proximity to a drive lane
truck is a rare occurrence. The frequency of occurrence is less than the assumptions used in the
original calibration studies (Nowak 1999). The largest frequency of occurrence is demonstrated
by the Tennessee IH-40 westbound (WB) data, where approximately 30% of traffic is accom-
panied by another vehicle located with a headway distance of less than 1,000 feet, and 1.8% of
traffic is accompanied by another vehicle located with a headway distance less than 20 feet. For
the most critical passing lane position for cross-frames (one directly behind another or zero
clear distance), the largest frequency of occurrence is demonstrated to be 0.03%. Recalling that
cross-frame forces generally reduce rapidly as a truck moves away from the cross-frame in the
Table 3-10. Summary of multiple presence statistics for adjacent lane loaded.
Frequency of Occurrence within Clear Distance Window, Relative to Drive Lane Annual
Truck Traffic (ATT) a
WIM Site
+/- 1,000 ft +/- 280 ft +/- 50 ft +/- 20 ft +/- 0 ft
Indiana
4.0% 2.1% 0.5% 0.3% 0.01%
US-31 NB
Indiana
4.2% 2.2% 0.5% 0.3% 0.00%
US-31 SB
Tennessee
16.0% 7.4% 1.7% 1.0% 0.02%
IH-40 EB
Tennessee
29.9% 13.2% 3.0% 1.8% 0.03%
IH-40 WB
Virginia
3.9% 2.0% 0.5% 0.3% 0.01%
US-29
a
Reference Figure 2-11 for illustration of positive and negative clear distances.

longitudinal direction, it is apparent that, for cross-frames, the largest frequency of occurrence
for passing vehicles occurring simultaneously is appreciably low (i.e., larger headway distances
generally do not result in superimposed cross-frame forces). Another observation is that the
distribution of the passing lane truck GVW appears to be bi-modal for several sites, with
the heavier truck mode being equal to or heavier than the drive-lane truck GVW. Sensitivity
studies were conducted by White (2020) that explicitly accounted for the cross-frame force
effects caused by passing lane vehicles, and the effects on the accumulated fatigue damage were
found to be negligible.
The WIM study performed in this research confirmed that the dual truck event initially
considered in the 7th Edition AASHTO LRFD is a rare occurrence. As such, the current load
criteria (i.e., a single design truck positioned in all longitudinal and transverse positions) is more
appropriate. This multiple presence study indicates that even when considered, the effects of
passing lane traffic on both the Fatigue I and Fatigue II limit state parameters are negligible.
A comprehensive multiple lane analysis using all available two-lane traffic records demonstrates
that the probability of a single truck record (regardless of GVW) being located in a critical
position for magnifying cross-frame force effects is exceptionally low—less than 0.02% for the
most critical location (i.e., the steering axle of one truck is located just behind the rear axles
of another truck in an adjacent lane) and less than 3% when the truck is located within 20 to
50 feet from the rear axle of another truck in an adjacent lane.
3.2.4 Major Outcomes

Based on the results of the Fatigue Loading Study presented in the preceding subsections,
there are several major conclusions that can be drawn with respect to Objectives (a) and (b) and
the five questions posed in the introduction of Section 3.2:
• Cross-frames are sensitive to transverse truck placement. For cross-frames near the maxi-
mum positive dead load moment region, the longitudinal load influence tends to be localized.
For cross-frames in straight or curved girder systems near skewed end or intermediate sup-
ports, the load influence response is significantly broadened. In general terms, the influence
of load position is highly variable and difficult to predict unless an influence-surface analysis
is conducted.
• Similarly, critical lane position also depends on a variety of parameters. In general, truck
passages along the outer edges of the deck (i.e., overhang loads) tend to maximize cross-
frame forces in skewed and curved bridges (i.e., load applies net torque on superstructure
which engages the cross-frames) but not in straight, normal bridges. Additionally, cross-
frame response is generally linearly dependent on the location of the applied load relative
to the centerline of the fascia girder; that is, wheel loads that are 3 feet outboard of the fascia
girder produce larger force demands on critical cross-frame members than loads that are
1 foot outboard.
• The WIM study confirmed that the dual truck event (i.e., multiple presence) initially
considered in the 2016 Interims to the 7th Edition AASHTO LRFD Specifications is a rare
occurrence. In terms of implementation of AASHTO LRFD, the results also indicate that the
current specification language in Article 3.6.1.4.3a (“a single design truck shall be positioned
transversely and longitudinally to maximize stress range at the detail under consideration”) is the
best method to ensure the critical load position for the governing cross-frame is considered
by the designer.
• The governing cross-frame in a given bridge depends on many variables, including girder
spacing, support skew, and lane positions. Bottom struts tend to be the governing cross-
frame member in most bridges, as cross-frames tend to act as composite “floorbeams” when
distributing applied live loads transversely to adjacent girders. Still, the response is likely too

variable to pinpoint which cross-frames are critical without some refined, influence-surface
analysis (i.e., to examine all possible truck positions and cross-frame force effects).
• The correlation between specific variables and cross-frame fatigue design forces was
explored. In general, increased skew and tighter bridge curvature tends to increase cross-
frame force demands in composite systems. These general rules-of-thumb can potentially be
utilized by engineers to mitigate cross-frame forces in the design phase of a project. In other
words, the engineer can potentially avoid an iterative “chase-your-tail” design solution by
adjusting other key parameters. For instance, using discontinuous, staggered cross-frame
layout is a practical and economical solution, whereas increasing the deck thickness or
decreasing girder spacing has significant impacts on the rest of the design.
• From the fatigue truck study, it was concluded that additional economies in the AASHTO
fatigue criteria are possible for cross-frame design. This is largely attributed to the sensi
tivity of cross-frame response to transverse load position. Current design criteria require
all possible lane positions be considered in accordance with AASHTO Article 3.6.1.4.3a,
regardless of design lanes or actual lane striping. A more realistic scenario is to consider only
drive lanes (i.e., lanes within the limits of the fascia girders, for which most truck traffic
traverses). Alternatively, reduction factors specific to cross-frames could be applied to
the existing load factors. Based on feedback received from the review panel, cross-frame-
specific load factors outlined in the subsequent bulleted item were explored as the preferred
approach.
• To maintain the current guidance in terms of truck position (i.e., the AASHTO fatigue
truck is located in every transverse position on the bridge deck between the vehicle barriers),
cross-frame-specific fatigue load factors were investigated to eliminate the source of conser-
vatism in the current load factors. The modified load factors developed in this study are
based on a comprehensive WIM study utilizing calibrated bridge models representative of
a variety of straight bridges with normal supports, straight bridges with skewed supports,
and horizontally curved bridges. Assuming the WIM records represent typical truck weights
throughout the country, the improved load factors reflect more realistic force effects experi-
enced by cross-frames in these types of bridges. Using published statistical data for resistance,
the reliability of using these improved load factors was investigated. The Monte Carlo
simulation using the available resistance data and cross-frame statistics summarized in this
report result in reliability indices as shown in Table 3-9 when using an adjusted load factor
of 0.65 applied to both of the current Fatigue I and II limit state load factors (i.e., a respec-
tive Fatigue I and II load factor for cross-frames of 1.14 and 0.52). Based on this analysis,
the resulting reliability indices for cross-frames using these adjusted load factors meets or
exceeds the assumed target reliability of unity (Modjeski and Masters 2015).
• For the analytical and experimental analyses conducted in this study, the number of design
stress cycles per truck passage (n in AASHTO LRFD Table 6.6.1.2.5-2) was taken as 1.0.
The proposed load factors above were subsequently calibrated using this n = 1.0 assumption.
• In implementing the reduced cross-frame-specific load factors proposed, it was observed
that load-induced fatigue problems are likely only an issue in moderately to heavily skewed
and/or curved bridges (i.e., Ic > 1, Is > 0.15, or Ic > 0.5 & Is > 0.1) in areas of significant truck
traffic volume. For bridges that do not fall into one of these categories, a refined 2D or 3D
analysis is not warranted. These limits are similar to those established in NCHRP Research
Report 725 (White et al. 2012), which focused primarily on analysis of noncomposite systems,
except that the skew index limit proposed in that study is lowered herein from 0.3 to 0.15.
A discussion on the appropriate analysis method is outlined in the subsequent sections.
These findings, which are further summarized in Chapter 4, served as the basis for many of
the proposed modifications to AASHTO LRFD. Refer to Appendix A for the proposed language
and commentary.

3.3 R-Factor Study (3D Analysis)

This section summarizes the results related to the R-Factor Study, which focuses on
addressing Objective (c) of NCHRP Project 12-113. More specifically, the limitations of the
traditional truss-element modeling approach for cross-frames in 3D models are investigated
in the context of composite, in-service bridges. The applicability of a proposed alternative
approach (eccentric-beam model) is also assessed for a variety of bridge geometries. These goals
were summarized by the following questions, which were initially posed Section 1.2:
• Is the current established R-factor (0.65AE), which was based on analytical and experimental
studies of a noncomposite system, appropriate for cross-frames in the composite condition?
• Are there alternative 3D modeling approaches for cross-frames?
These questions are systematically addressed throughout this section of the report. Section 3.3.1
summarizes key findings from a panel-level study, which analytically examined the response of
noncomposite cross-frame panels to various load-induced deformation patterns anticipated in
in-service bridges. Section 3.3.2 presents the results of a 3D FEA parametric study that inves-
tigated the response of cross-frames in composite bridge systems. The accuracy of the stiffness
modification factors and the eccentric-beam approach are assessed in terms of predicted cross-
frame force effects for a variety of bridge geometries. Lastly, Section 3.3.3 summarizes the key
findings from the study.
3.3.1 Panel-Level Studies (Noncomposite)

As noted in Section 2.4.4.1, a parametric study was performed to evaluate the stiffness response
of different X-type cross-frame configurations and connection details due to various deforma-
tion patterns in bridge girders, including equal rotation, differential vertical displacement, and
differential rotation. In general, stiffness modification (R) factors were derived by comparing the
computed torsional stiffness response of a benchmark shell-element model and the response of
the corresponding truss-element model. This process was repeated for a variety of cross-frame
panels and connections (approximately 2,000 unique cases).
Figure 3-21 summarizes the computational results. The computed R-factor is plotted
separately for the given loading condition. Each data point is grouped by angle leg thickness
(1⁄4-inch up to 5⁄8-inch) to highlight the relative impact of the parameters on the results.
Within each grouping, various angle leg lengths (3 inches and 4 inches) and girder spacing-
to-panel height ratios are grouped and presented. Figure 3-21 provides specific callouts to
demonstrate how the various parameters are presented. Additionally, the results presented
correspond to a constant gusset plate thickness of 0.5 inches only. However, note that an
increase in gusset plate thickness has been shown to generally produce higher modification
factors (i.e., the increase in out-of-plane flexural stiffness influences the overall stiffness more
than the additional eccentricity introduced).
A key emphasis of the panel-level analytical studies, as previously introduced in Section 2.4,
was related to the orientation of the cross-frame diagonals relative to the gusset and connection
plates. As such, the differential-rotation deformation pattern was considered with two distinct
cases similar to those schematically shown in Figure 2-16. These cases are identified as “same
side” connections and “opposite side” connections. The same side case corresponds to the
loading condition in which only the diagonal that frames into the same face of the gusset as
the strut is engaged; the opposite side case corresponds to the loading condition in which the
diagonal that frames into the opposite face of the gusset is engaged. The other loading con-
ditions, equal rotation and differential vertical displacement, engage both diagonal members
equally such that the resulting effects are not pronounced in the results.

1
0.9
0.8
0.7
Stiffness Modification Factor, R
0.6
0.5
0.4
Equal Rotation Differential Vertical Displacement
0.3
1
0.9
0.8
0.7
0.6
0.5
0.4
Differential Rotation – Same Side Differential Rotation – Opposite Side
0.3
1/4 3/8 1/2 5/8 1/4 3/8 1/2 5/8
Angle Thickness (in)
Figure 3-21. Computed stiffness modification factors for various

loading conditions and angle thicknesses.
From Figure 3-21, the following observations can be made with respect to the panel-level
study:
• In general, the parameters of the single-angle section (thickness, leg width) and the aspect
ratio of the panel (girder spacing, panel height) affect the computed modification factor, R.
The R-factor equation presented in Battistini et al. (2016) shows that increasing the S/hb ratio
(girder spacing-to-panel height), –y (distance between the face of the connected leg and the
neutral axis of the angle section), and t (angle leg thickness) results in a decreased R-factor for
the equal rotation condition. Similar trends for composite systems were generally observed
from the results of the studies described in the next section.
• Significant scatter in the results is observed for the differential-rotation case depending on
which cross-frame diagonal is engaged. The same side connections generally result in a more
flexible response due to the additive effect of the end connection eccentricities, whereas the
opposite side connections are much stiffer.
• Even for parameters dictated by designers (e.g., angle thickness, angle leg width, girder
spacing, panel height, gusset thickness), there can be substantial variability in the R-factor
response. Additional variability is also introduced by parameters not controlled by designers
(e.g., deformation pattern and the force distribution due to truck position).
Additional spot checks were performed to investigate the effects of different connection
plate dimensions (i.e., slenderness ratios) and K-type cross-frames. In brief, it was evident
that the connection plate slenderness had minimal effect on the computed R-factor for all
loading conditions evaluated. In terms of the K-type cross-frames, the “same and opposite
side” effects were not observed given that diagonals in K-frames typically connect to the
same face of the corresponding gussets. As such, it can be concluded that K-frames are not as
sensitive to the deformation pattern given the symmetry of the end connections. For a more
thorough review of these spot check analyses, refer to Appendix F.
Given the variability and uncertainty in the response due to different geometries and load-
ing conditions, developing a closed-form, general-use solution similar to that suggested in

Battistini et al. (2016) would be challenging and impractical for design. From a simplicity stand-
point, assigning a single modification factor for each cross-frame member, similar to the current
approach in AASHTO LRFD, is desirable. This is particularly true since the orientation of the
cross-frame connections (i.e., increasing or decreasing eccentricities) may not be known during
the design process. With that in mind, the system-level studies in the next subsection evaluate
the most appropriate modification factor to assign for the majority of bridge superstructures in
service. An independent parametric study that quantifies the impact of an assumed R-factor on
the cross-frame force prediction in composite bridge systems is outlined.
3.3.2 System-Level Studies (Composite)

Two independent parametric studies were performed as part of this system-level study
as outlined in Section 2.4.4. One study, consisting of only straight and normal bridges, was
performed to emphasize the influence of connection and gusset plate dimensions. The other
study was conducted to investigate the effects of support skew and horizontal curvature on the
R-factor and eccentric-beam approaches. Given that cross-frame force effects are more critical
in skewed and/or curved bridge systems, the results presented herein focus on the second study
only. For an extended overview of the entire data set, refer to Appendix F.
In general, multiple iterations of the same 3D bridge model were developed, and the cross-
frame force results were subsequently compared. A rigorous shell-element model served as the
benchmark model, and various truss-element and eccentric-beam models were produced to
assess the accuracy and potential limitations of these simplified methods.
Before discussing the implications of employing the proposed eccentric-beam approach to
modeling cross-frames, Figure 3-22 summarizes a sample set of results for the truss-element,
R-factor approach. In this figure, the design force range for each model iteration of a sample
bridge is reported (based on the AASHTO fatigue truck traversing in the critical lane). Along
the x-axis, the results for the various truss-element models are considered as a function of the
applied modification factor, R = {0.5, 0.6, 0.7, 0.8, 0.9, and 1.0}. Along the y-axis, the Fsimplified /Fshell
ratio is plotted.
A value of unity indicates that the assigned R-factor resulted in perfect agreement with the
shell-element model. Values above unity imply that the truss-element model is too stiff (i.e.,
the truss member attracts more force than the shell-element model) and that the R-factor
should be reduced. This case represents a conservative design assumption with respect to the
fatigue limit state (i.e., the analysis model produces excessive design forces) but is unconser-
vative from the perspective of stability bracing or torsional behavior in a curved girder. The
opposite is true for values below unity. In the callouts on Figure 3-22, note that the “exact”
R-factor, where the line intersects Fsimplified /Fshell = 1.0, is included for each member presented.
For example, an R of 0.73 yields identical behavior between the shell and truss models for the
bottom strut in the 1/2-inch gusset plate case.
Figure 3-22 presents this data for three select cross-frame members in the sample bridge
model. The select cross-frames are as follows: (i) a bottom strut near the maximum positive
dead load moment region, (ii) a top strut from the same cross-frame panel, and (iii) a diagonal
near the intermediate skewed support. The figure also differentiates the models by gusset plate
thickness (1⁄2-inch or 1-inch thick). Note that the gusset plate thickness only affects the shell-
element model directly, as the truss-element models do not explicitly represent gusset plates.
Thus, only the denominator in the Fsimplified /Fshell ratio is impacted between the top (1⁄2-inch thick)
and the bottom (1-inch thick) plots in Figure 3-22.
For reference, a sketch of the framing plan and typical cross-section are provided (not to
scale); the skewed diaphragms at the end and intermediate supports are included in the model

Key:
3
1 2
1.6
1/2-inch Gusset
1.4
Diagonal 1 (R = 0.63)
1.2
1
Fsimplified / Fshell (Critical Loading)
Bottom Strut (R = 0.73)

0.8 Top Strut (R = 0.63)
0.6
0.4
1.60.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1
1-inch Gusset
1.4
1.2 Top Strut (R = 0.56)
1
Bottom Strut (R = 0.95)
0.8
Diagonal 1 (R = 0.70)
0.6
0.4
0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1
Assigned R-Factor (Truss Model)
Figure 3-22. Example data from two-span, continuous skewed bridge

demonstrating the predicted cross-frame design force range from the
truss-element approach (using various R-factors) and the shell-element
approach.
but neglected in the figures for clarity. For more information regarding the pertinent parameters
of the superstructure, refer to Appendix F.
The following observations can be made from Figure 3-22:
• The cross-frame force effect is not 1:1 dependent on the stiffness modification due to the
high degree of indeterminacy in the bridge system. In other words, a 10% reduction in the
R-factor does not result in a 10% reduction in the estimated cross-frame force.
• Rather, a 10% reduction in the R-factor generally results in approximately a 5% reduction
on average in the estimated cross-frame force. That relationship typically holds true for
modification factors above 0.6. Below R = 0.6, the rate at which the Fsimplified /Fshell approaches
zero increases. Ultimately, a modification factor of zero results in Fsimplified /Fshell = 0, despite
not being shown in the figure. This assumed 2:1 relationship (x% reduction in R ≈ x/2%
reduction in force) has design implications. For example, if an acceptable level of error in the
predicted cross-frame force is established as 10%, the R-factor likely needs to be within 20%
of the exact value (recall that it is conservative from a fatigue perspective to assign higher
R-factors).
• Even for cross-frame members of similar geometries and eccentricities, the “exact” R-factor
for these critical loading conditions can vary. For example, the exact R-factor for the bottom
strut is 0.73 compared to 0.63 for the top strut in the same cross-frame panel. This variability
highlights the impact of load position and deformation patterns on the assumed stiffness.
• The gusset plate thickness also impacts the response of the cross-frame in the shell-element
model. As the gusset plate thickness increases, the stiffness of the cross-frame tends to also
increase. As noted earlier, although the eccentricity is larger with a thicker gusset plate, the

bending stiffness of the plate increases. The reason for this is that the eccentricity is a linear
function while the bending stiffness changes as a cubic function. As such, the exact R-factor
for the bridge with 1-inch thick gusset plates tends to exceed the exact R-factor for the
bridge with 1⁄2-inch thick gussets. For example, the exact R-factor for the diagonal member
evaluated was 0.63 for 1⁄2-inch gussets and 0.70 for 1-inch gussets.
• In general, the sample results presented in Figure 3-22 are representative of the full data set
in terms of the observed variability and trends.
Figure 3-22 represents the results of just one bridge in the sample set. Rather than evaluate
the effectiveness of the R-factor approach for every cross-frame member and every transverse
lane passage, it is more important to just focus on the governing cross-frame member in every
bridge (i.e., the cross-frame with the maximum force range). In this manner, more generalized
observations about the stiffness modification factors can be made. Note that a similar approach
was taken when compiling the data obtained from the Fatigue Loading Study.
Thus, for each iteration of a given bridge model, Fsimplified /Fshell ratios were computed and
compiled in the form of box-and-whiskers plots. The results of the eccentric-beam approach
(“refined model” only; refer to Figure 2-15) were also included to examine its potential benefits.
Note that, while the results herein focus on the refined eccentric-beam model only, an examina-
tion of the simplified versions is presented later in this section.
For Figure 3-23, the box-and-whiskers components are organized by the assigned R-factor
in the truss-element model, as well as the eccentric-beam model (which is independent of an
assigned R-factor). Therefore, each box-and-whiskers represents a data set of Fsimplified /Fshell ratios
corresponding to the truss-element models with R = {0.5, 0.6, 0.7, 0.8, 0.9, and 1.0} and the
eccentric-beam model. By presenting the box-and-whiskers of the results for various R-factors
side-by-side, the research team can evaluate which fixed factor generally produces the most
accurate representations of the “true” cross-frames stiffness (i.e., the shell-element model).
Similarly, the variability of the R-factor approach can be compared to the eccentric-beam
approach, which more accurately represents the stiffness of the cross-frame and its connections.
Statistical parameters such as the mean, minimum, and maximum values of the data set are
also provided. For the R = 0.5 example, the respective mean and maximum Fsimplified /Fshell ratios
were 0.82 and 0.98, which indicates that assigned R = 0.5 always underpredicted the design
force when compared to the benchmark shell-element model.
1.4
1/2-inch Gusset
Fsimplified / Fshell (Governing Member)
1.3
1.2
1.1
1.0
0.9
25th Percentile
0.8
0.7
0.6
0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.0 Ecc.
Beam
Assigned R-Factor (Truss Model)
Figure 3-23. Box-and-whiskers plot indicating the overall level

of accuracy for the eccentric-beam modeling approach compared
to the truss-element approach (1/2-inch gusset plate).

The figure only presents the results related to the bridges analyzed with 1⁄2-inch-thick gusset
plates. However, bridges with 1-inch-thick gusset plates were evaluated, and the results are
subsequently provided in Appendix F. In addition to the box-and-whiskers, a line graph
representing the 25th percentile of each data set is provided, whose meaning is outlined below.
Similar to the sample bridge in Figure 3-22, the following observations can be made about
the stiffness modification (R) factor approach with respect to the full data set presented in
Figure 3-23:
• As the assigned R-factor increases, the design force in the truss-element model increases,
thereby increasing the Fsimplified /Fshell ratio. In other words, the R-factor and Fsimplified /Fshell ratio
are positively correlated.
• Despite the positive correlation, there is still significant variability in the observed response.
The accuracy of the truss-element model in terms of predicted cross-frame forces is not
only a function of the assigned modification factor but also the bridge geometry and loading
conditions. For example, the respective maximum and minimum Fsimplified /Fshell ratios for the
R = 0.5 case are 0.98 and 0.73.
• Although not explicitly shown in this report, the Fsimplified /Fshell ratio decreases as the gusset plate
thickness increases. This is attributed to the fact that an increased gusset thickness results in a
stiffer cross-frame, which then attracts more force to the shell-element model.
From the observations above, it is apparent that developing an expression that precisely
predicts the exact R-factor for any general bridge model (i.e., the R-factor for which Fsimplified /Fshell
always achieves unity) is not practical or feasible. The exact R-factor is a function of many
parameters, including bridge geometry, loading conditions, cross-frame details (e.g., gusset
plate thickness), and member type (e.g., bottom strut versus diagonal). Thus, it is more appro-
priate to assign a uniform R-factor to all cross-frame members that statistically represents the
majority of bridge conditions, similar to the R = 0.65 approach currently adopted in AASHTO
LRFD that was developed for the construction condition of the bridge.
Based on these results, there are three potential approaches for handling the variable response
of the R-factor approach in terms of AASHTO LRFD. These are listed below from most
sophisticated to most simplistic as well as potential pros and cons to each approach:
• Option 1: Develop a member-specific and bridge-specific expression that provides a precise
R-factor. Given the variability in the response and the uncertain nature of live loads on bridge
structures, this approach is neither feasible nor practical for implementation in AASHTO.
The expression would inevitably be extremely complex and would still likely produce
scattered results.
• Option 2: Develop an R-factor that is a direct function of the gusset plate thickness. Rather
than develop an expression that explicitly considers all pertinent bridge and cross-frame
parameters (e.g., girder spacing, angle thickness), this approach simply focuses on gusset
plate thickness, which has been shown to substantially impact the exact R-factor. This
procedure could be easily implemented as an equation or a table. However, given that the
gusset plate thickness may be variable during the design process, this approach potentially
lends itself to an iterative analysis procedure. Even with this extra level of refinement, the
results could still be variable.
• Option 3: Develop a single R-factor that generally represents the behavior of all cross-frame
members in a conservative manner (similar to the current 0.65 approach in the 9th Edition AASHTO
LRFD that was developed for the construction condition). As demonstrated with all previous
results, pinpointing an exact R-factor is very difficult given its inherent dependence on the
loading conditions and the associated deformation patterns. Regardless of how the cross-
frames are modeled and R-factors are assigned, some error in cross-frame force predictions

is to be expected when compared to the shell-element modeling approach. With that said,
assigning a single R-factor for all cross-frame members, regardless of connection details,
bridge geometry, and loading conditions, is a viable option.
After weighing the options, it was elected to recommend the use of a single R-factor
(Option 3) similar to the current approach taken in AASHTO Article C4.6.3.3.4. With that in
mind, it is also not appropriate to ensure every bridge model results in conservative estimates
of cross-frame fatigue forces (Fsimplified /Fshell ≥ 1). Instead, the research team investigated different
statistical parameters but ultimately elected to base the proposed modification factor on the
25th percentile value of the data sets. In other words, the intersection of the 25th percentile line
and Fsimplified /Fshell = 1.0 represents the assigned R-factor that provides a conservative estimate
of cross-frame design forces for 75% of the bridges. For the 25% of the bridges that result in
unconservative estimates, it has been shown that the truss-element model is still within 10%
of the rigorous shell-element model, which was deemed reasonable by the research team.
Using that metric, the most appropriate R-factor to assign for the bridges with 1⁄2-inch gussets
is 0.75. As such, in addition to the stiffness modification factor of 0.65 that was derived for a
noncomposite system during construction, a second factor of 0.75 is proposed for structural
analysis of the composite condition.
Table 3-11 tabulates the results outlined above, as well as additional data corresponding to
different gusset plate thickness values and statistical parameters. For instance, if the in-service
R-factor is based on the 50th percentile (i.e., mean) of the data set with 1⁄2-inch-thick gusset
plate, then R = 0.66 is more appropriate, which is approximately the same as the established
modification factor in AASHTO LRFD (R = 0.65).
Although the R-factor approach provides a method of correcting truss-element models for
the reduction in stiffness due to eccentric connections, some designers or software producers
may opt to make use of a model that directly considers the connection eccentricity. In terms of
the proposed eccentric-beam results in Figure 3-23, it was observed that the refined eccentric-
beam models predict the cross-frame fatigue design forces with reasonable levels of accuracy.
For the 1⁄2-inch gusset case, the Fsimplified /Fshell ratio is 0.96 on average with a maximum of 1.03 and
a minimum of 0.86. In other words, the eccentric-beam produces results within 4% of the
shell-element model on average.
As outlined above, the results presented in Figure 3-23 are based on the “refined” eccentric-
beam model only. To illustrate the impacts of the more simplified eccentric-beam models,
Figure 3-24 presents sample influence-line results from the same two-span, continuous skewed
bridge from Figure 3-22. In this figure, two different responses are examined: (i) the axial force
in the bottom strut member near the skewed intermediate support when the AASHTO fatigue
truck traverses the bridge along the barrier and (ii) the axial force in the bottom strut member
(interior bay) near the maximum positive dead load moment region when the AASHTO fatigue
truck traverses the bridge along its centerline.
Table 3-11. Summary of R-factor results from system-level

parametric study.
Appropriate R-factor
Gusset thickness [in]
Mean 25th Percentile
1/4 0.60 0.65

1/2 0.66 0.75
3/4 0.70 0.81

BS
BS Case 1
Case 2
9 CF Model FRa (kips) % Error

Shell-Elements (Control) 11.0 --
6 Truss-Elements; R={0.5, 1.0} {8.7, 13.3} {-21%, +21%}
Ecc. Beam; Refined 10.4 -6% R = 1.0
3 Ecc. Beam; Simplified 10.7 -3%
Ecc. Beam; Angle (Fixed Ends) 12.8 +16%
Ecc. Beam; Angle (Pinned Ends) 6.8 -38% R = 0.5
0
-3
-6
Axial Force (kips)
Case 1
-9
0 210 420
6 CF Model FRa (kips) % Error
R = 1.0
Shell-Elements (Control) 4.8 --
Truss-Elements; R={0.5, 1.0} {3.9, 5.6} {-21%, +14%}
Ecc. Beam; Refined 4.6 -6%
3 Ecc. Beam; Simplified 4.1 -16%
Ecc. Beam; Angle (Fixed Ends) 4.8 -2%
R = 0.5 Ecc. Beam; Angle (Pinned Ends) 3.2 -35%
aFR = Force Range

Case 2
-3
0 210 420
Figure 3-24. Example data demonstrating the predicted cross-frame force

range for the various eccentric-beam models.
For each cross-frame response, several different results are presented, including the shell-
element model (which serves as the control), the truss-element model range (i.e., R = 0.5 and
R = 1.0), the refined eccentric-beam model, the simplified eccentric-beam model, and the
truss-only eccentric-beam models (both fixed-end and pinned-end). The various responses are
superimposed on the same graph. Additionally, the total force ranges and the percent error
relative to the control model are tabulated for reference. Note the scale along the vertical axis
is different for the top and bottom plots in Figure 3-24. Refer to Appendix F for a detailed
description of each eccentric-beam modeling approach.
The following observations can be made from the sample data set presented in Figure 3-24:
• Similar to the results presented above, the truss-element model approach has significant
variability depending on which R-factor is assigned. Note that a modification factor of
approximately 0.75 would produce the most accurate results in both cases shown in
Figure 3-24, which is consistent with the conclusions of the R-Factor Study presented in
the previous section.
• In both loading scenarios, the refined eccentric-beam model produces relatively accurate
results when compared to the shell-element model (6% conservative). This is a similar
observation to the results of the parametric study presented above. For instance, these two

data points are close to the mean of the eccentric-beam data set (box-and-whiskers) shown
in Figure 3-23 (i.e., a Fsimplified /Fshell ratio of 0.94 compared to 0.96 in Figure 3-23).
• The simplified eccentric-beam model, which neglects the individual eccentric offsets caused
by the connection and gusset plates, has more variability but generally produces reasonably
accurate and conservative results. This approach is much simpler to employ in practice and
requires much less information about the cross-frame connection details.
• The angle-only models are largely dependent on the assumed end restraints. When the beam
ends are fixed rotationally to the girder web, the stiffness of the cross-frame panel tends
to be overestimated, which results in slightly overconservative force effects (but still less
conservative than the truss-element model assuming R = 1.0). Despite the eccentric offset
being explicitly modeled, the out-of-plane flexural rigidity is greatly overestimated at the end
connections by using the angle section properties for the full length of the modeled element.
• When the beam ends are pinned (to account for the rotational flexibility of the connection
plates), the overall cross-frame stiffness tends to be vastly underestimated (even more under-
estimated than the truss-element model with R = 0.5). Thus, the cross-frame force effects are
significantly unconservative with respect to the shell-element model. It can be observed from
this sample data that an eccentric-beam approach, for which rotation is released at its ends,
is not an appropriate 3D modeling technique for cross-frames.
• In review of the angle-only models, the “real” rotational restraint at the ends of the beam
elements is somewhere between the idealized fixed and pinned scenario. Although not
studied here, a rotational spring with finite stiffness could be explored as an alternative as well.
The most significant observation from the sample results in Figure 3-24 is that accurately
representing the out-of-plane flexibility associated with the connection and gusset plates is
vital for eccentric-beam models. The angle-only models generally do a poor job of representing
that end flexibility (assuming idealized fixed- or pinned-end conditions), whereas the simplified
and refined models showed much more promise.
In general, the eccentric-beam approaches eliminate several sources of uncertainty associated
with the R-factor approach (e.g., eccentric distances, relative stiffness of gusset and connection
plates, effect of load position on the deformed shape). However, modeling a cross-frame based
on the refined methodology requires knowledge of the gusset and connection plate thicknesses
prior to designing those elements. This potentially lends itself to an iterative process where the
engineer starts with an assumed geometry (likely based on standard DOT details) and updates
the model based on project-specific design decisions. This iterative process, though, affects
the shell-element and refined eccentric-beam approaches equally. In contrast, the simplified
methodology, although less precise and accurate, reasonably approximates the flexibility of the
connection plate without having full detailed dimensions of the cross-frame connections.
Based on these perceived advantages, this modeling technique has potential to serve as
an alternative for bridge designers and commercial design software packages. As a reference,
the current commentary language in AASHTO LRFD (C4.6.3.3.4) that discusses the use of
the R-factor is provided herein:
“In addition, the axial rigidity of single-angle members and flange-connected tee-section cross-frame
members is reduced due to end connection eccentricities (Wang et al., 2012). In lieu of a more accurate
analysis, (AE)eq of equal leg single angles, unequal leg single angles connected to the long leg, and flange-
connected tee-section members may be taken as 0.65AE.”
As underlined in the provision above, the eccentric-beam approaches (namely the refined
and simplified approaches) qualify as a “more accurate analysis” but do not require bridge
designers to assemble labor-intensive, 3D FEA models with cross-frames modeled as shell
elements. The overall conclusions with regards to this proposed analysis procedure are provided
in Section 3.3.3 below.


Based on the results of the R-Factor Study presented in the preceding subsections, there are
three major conclusions that can be drawn with respect to simplified 3D analysis methods for
cross-frame design:
• The appropriate R-factor to be assigned in truss-element models is largely a function of
bridge geometry, cross-frame details, and uncertain loading conditions. Consequently,
considerable scatter was observed in all phases of the analytical studies. Still, the R-factor
approach is a simple solution to a complex problem that produces reasonably accurate approxi-
mations of the actual cross-frame stiffness. Considering that many designers often prefer
simple alternatives over sophisticated refined analyses, it serves an important role in AASHTO
LRFD guidance moving forward.
• In terms of implementation into AASHTO LRFD, a reasonable approach considering the
numerous uncertainties is the use of a single R-factor, similar to the current approach adopted
into the specifications. Rather than use the current AASHTO recommendation of R = 0.65
that was developed based upon the construction condition, R = 0.75 has been shown to pro-
duce more accurate results in composite systems when compared to benchmark solutions. As
such, it is recommended to propose two separate factors, Rcon for construction stages (0.65)
and Rser for in-service conditions (0.75). However, it should be noted that additional statistical
parameters have been investigated that result in different potential modification factors.
• The proposed eccentric-beam model represents an approach that is slightly more refined
than the conventional use of pin-ended truss elements, but less complex than modeling
cross-frames with shell elements. As long as the out-of-plane connection plate flexibility is
properly considered, the results demonstrated that the proposed method improves repeat-
ability and reliability of 3D models by eliminating several sources of uncertainty associated
with the R-factor approach. Although the R-factor approach serves a vital purpose in practice
due to its ease of use and familiarity, the proposed eccentric-beam method offers another
approach to engineers seeking a more refined solution.
and commentary.
3.4 Commercial Design Software Study (2D Analysis)

This section summarizes the major findings of the Commercial Design Software Study that
was previously introduced and outlined in Section 2.5. The study, which specifically addresses
Objective (d) of NCHRP Project 12-113, investigates the common 3D and 2D analysis tech-
niques employed by commercial design software programs with respect to estimating cross-frame
force effects. This objective is otherwise summarized by the following questions that were
initially posed in Section 1.2:
• What are the limitations of simplified 2D analysis techniques commonly used by popular
commercial design software programs in terms of predicting cross-frame force effects in
composite systems?
• Are there methods available to improve these simplified techniques?
By comparing the performance of 3D models, 2D PEB models, and 2D grillage models, these
questions are systematically addressed throughout this section of the report. Similar to the
Fatigue Loading Study, an abundance of cross-frame data was obtained from the Commercial
Design Software Study. To simplify the results and discussion for the reader, a similar outline
with respect to data presentation is followed for the discussion of all models. Representative

data are first shown in Sections 3.4.1 and 3.4.2. These sections address different aspects of the
results. Section 3.4.1 presents sample influence-line results to introduce the general limita-
tions of 2D analysis methods, and Section 3.4.2 investigates the influence of composite action
on the accuracy of these 2D models. Section 3.4.3 compiles and summarizes the results for the
full data set. Section 3.4.4 proposes an alternative postprocessing procedure for 2D PEB analyses,
and Section 3.4.5 summarizes the major outcomes.
3.4.1 Sample Influence-Line Results

As a starting point, Figure 3-25 compares the predicted cross-frame response for various
analysis methods under a specified moving load. This sample highlights key trends and obser
vations that are further validated in subsequent sections.
Figure 3-25 represents a straight bridge with normal supports. Before investigating more
complex framing systems, it is prudent to assess these simplified 2D analysis methods on a
simple structure. Simplified 2D analyses have been generally perceived to produce erroneous
cross-frame results in heavily curved bridges and near skewed supports, because they are unable
1 2
8 Key:
Case 1
[Approach; Equiv. Beam]
6 A Control (Abaqus)
B 3-D
4
A/B D-2 C-1 2-D PEB; Shear
2 C-3 D-3 C-2 2-D PEB;
D-1
Flexural
C-2
0 C-3 2-D PEB;
C-1 Timoshenko
-2
D-1 2-D Grid;
2 0 A/B
240
Flexural
D-3
0 D-2 2-D Grid;
Axial Force (kips)
C-1 Timoshenko
D-1
-2 C-2 C-3 D-3 2-D Grid;
D-2 Composite
-4 Note: Framing plan
sketch above not to scale
-6
Case 2
-8
4 0 240
Case 3
2 A/B
D-1/D-2/D-3
0
C-2/C-3
-2 C-1
-4
-6
0 240
Figure 3-25. Comparison of analysis methods for three select

cross-frame members in a straight bridge with normal supports.

to capture the torsional response of the bridge and the complex load paths through the structure
in each case. These geometric effects are addressed in subsequent sections.
The axial-force response of the three select cross-frame locations is shown: (i) a bottom strut
near midspan referred to as Case 1, (ii) a top strut from the same cross-frame panel referred
to as Case 2, and (iii) a diagonal near the support referred to as Case 3. This is demonstrated
schematically in the framing plan sketch in Figure 3-25.
The AASHTO fatigue truck was moved longitudinally along the centerline of the bridge as is
depicted. Figure 3-25 plots the axial-force response of the respective cross-frame members as
a function of the truck position along the bridge length, measured from the front axle to the
start of the bridge. These graphs are similar to the influence-line plots presented in Figure 3-24.
This process was repeated for several variations of the same bridge model including the 3D
validated approach in Abaqus (control) and the following methods in the commercial design
software program Software A (recall that Software A has the ability to create either a 3D model
or a variety of 2D models including PEB and grillage models):
• 3D model,
• 2D PEB model with equivalent cross-frame beams based on the shear-analogy approach,
• 2D PEB model with equivalent cross-frame beams based on the flexural-analogy approach,
• Improved 2D PEB model with equivalent cross-frame beams based on the Timoshenko
approach,
• 2D grid/grillage model with equivalent cross-frame beams based on the flexural-analogy
approach,
• 2D grid/grillage model with equivalent cross-frame beams based on the Timoshenko
approach, and
• Improved 2D grid/grillage model with equivalent cross-frame beams considering the
transverse stiffness of the concrete deck and the modified postprocessing procedure per
Figure 2-17 (abbreviated as 2D Grid Composite in the Figure 3-25).
The specifics of these 2D modeling approaches (e.g., equivalent cross-frame beam analogies)
were outlined in Section 2.5. Note that, for the PEB and grillage models, equal distribution
of cross-frame shear was assumed between top and bottom nodes when postprocessing the
respective analysis results. The implications of using a different distribution assumption are
addressed in subsequent sections. For more information regarding the overall geometry and
cross-frame layout of this sample bridge as well as additional influence-line results, refer to
Appendix F.
From Figure 3-25, the following observations can be made when comparing the results of a
specific cross-frame location across each bridge model:
Case 1 (bottom strut near the maximum positive dead load moment region):
• The 3D commercial design software model produces excellent results when compared to the
validated 3D FEA model in Abaqus (i.e., estimated force range within a few percent). Given
that the commercial design software modeling approach is nearly identical to that of the
control model, this trend was anticipated. This also indicates that the connection point of
the cross-frame members (i.e., into a shared node on the web-flange junction versus at an
offset distance along the web) is less impactful for 3D model results.
• The 2D PEB results were not significantly influenced by the equivalent beam method employed
for these composite conditions. In general, these models produced accurate estimates of
bottom strut forces when compared to the refined 3D models.
• The 2D grid models generally did a poor job at predicting the cross-frame forces, except when
the Timoshenko beam approach was implemented.

Case 2 (top strut near the maximum positive dead load moment region):
• Similar to Case 1, the 3D commercial design software model generally produces excellent
results when compared to the validated 3D FEA model in Abaqus.
• For the 2D analyses that do not consider the contributions of the deck when postprocessing
the analysis results, the top strut results are equal in magnitude and opposite in sign to those
results of Case 1. This is a function of the simplified postprocessing used when converting
the equivalent beam back into a truss. As illustrated in this example, this simplification
generally results in erroneous results for top strut forces.
• This error is corrected in the 2D grillage model using cross-frame beams with equivalent
section properties to account for the contributions of the deck. Albeit the correct sign was
obtained, the magnitudes are still generally overestimated.
Case 3 (diagonal near the intermediate/end support):
• In general, the observations for Case 3 are similar to those for Case 1 except that the 2D
approaches consistently underpredicted the cross-frame force effects.
In reviewing data beyond these sample results (specifically for bridges with X-type cross-
frames), it was evident that the predicted response of bottom strut member forces in 2D models
was much more accurate than the predicted response of diagonal or top strut members. This
behavior was apparent in both simple and complex bridge geometries. In other words, the
variability observed in the response was not only a function of the bridge geometry (i.e., straight
versus curved, normal supports versus skewed supports) but also the cross-frame member type
(i.e., bottom strut versus diagonals).
The variability associated with bridge geometry is attributed more to analysis error (i.e., how
the analysis determines end shears and moments on the equivalent cross-frame beam), whereas
variability associated with member type is attributed more to postprocessing error (i.e., how
those end shears and moments are converted into cross-frame member forces). With that in
mind, Section 3.4.2 examines this behavior in the context of noncomposite and composite sys-
tems to determine the root cause of these discrepancies.
3.4.2 Postprocessing Error

With general trends laid out in Section 3.4.1, this section explores the error associated with
the postprocessing assumptions inherent to 2D modeling approaches. Analyses were performed
for noncomposite and composite systems to better understand how a composite deck affects
the distribution of forces in cross-frame systems. Namely, the research team sought answers to
the following questions: are the assumptions inherent to noncomposite systems (e.g., equal shear
distribution to top and bottom nodes) valid for composite systems; and, if not, how does an engineer
make adjustments to correct those assumptions?
Before examining more complex bridge systems, Figure 3-26 presents the results for the same
single-span, straight bridge with normal supports presented in Figure 3-25. By starting with a
simple bridge geometry, the postprocessing procedures can be isolated from any complexities
introduced with support skews and horizontal curvature. The response of a midspan cross-frame
panel is evaluated for a static loading condition (i.e., applying a 100-kip load just to the left of
the cross-frame panel of interest). Note that an additional static load case was considered in
this study, and those results are summarized in Appendix F. In total, four different analysis
methods are compared:
• 3D Software A model of a composite system (top right quadrant of Figure 3-26),
• 3D Software A model of a noncomposite system (bottom right quadrant),

100 k
2D PEB (Composite System) 3D (Composite System)
2.6 k
756 k-in 258 k-in
4.2 k 4.2 k
8.4 k
12.6 k 8.5 k 4.3 k 0.5 k 2.6 k 12.3 k
2.1 k 2.1 k 1.0 k 4.7 k

60” 60”
2.1 k 2.1 k 4.7 k 1.0 k
8.5 k 8.4 k
12.6 k 4.3 k 17.7 k 6.3 k
120” 120”
2D PEB (Noncomposite System) 3D (Noncomposite System)
21.1 k
2090 k-in 323 k-in
14.7 k 14.7 k
20.8 k
34.8 k 20.1 k 5.4 k 35.6 k 21.1 k 6.3 k
7.4 k 7.4 k 7.2 k 7.4 k
7.4 k 7.4 k 7.4 k 7.2 k

20.1 k 20.8 k
34.8 k 5.4 k 35.5 k 6.6 k
Figure 3-26. Sample 2D PEB and 3D cross-frame results for straight

bridge with normal supports (noncomposite and composite) under static
load (k = kips, k-in = kip inches).
• 2D PEB model of a composite system utilizing the Timoshenko beam approach for equiva-
lent cross-frame beam properties and the equivalent torsion constant (Jeq) for girder beam
properties (top left quadrant),
• 2D PEB model of a noncomposite system utilizing the Timoshenko beam approach for
equivalent cross-frame beam properties and the equivalent torsion constant (Jeq) for girder
beam properties (bottom left quadrant).
Given that the preliminary results presented in Figure 3-25 demonstrated the benefits of
PEB models over grillage models, the results in this section are focused on just PEB methods.
For the 3D analyses (right side of figure), the axial-force output from the cross-frame members
is presented, followed by the nodal forces required to equilibrate the panel. For the 2D PEB
analyses (left side of figure), the shear and moments produced from the equivalent beam
analysis are presented, followed by the cross-frame force effects computed based on the common
postprocessing assumptions previously outlined.

For each figure, there are three metrics by which the results can be compared, as listed below:
1. The total shear force acting on the cross-frame panel relative to the shear force in the concrete
deck: This value provides an indication of how accurate the equivalent beam properties
in the 2D model are. For instance, if the total shear force determined from the 3D model
exceeds that of the 2D PEB model, then the stiffness of the equivalent cross-frame is likely
underestimated. The opposite would be true if the total shear force of the 2D PEB model
exceeds the 3D model.
2. The distribution of that total shear force to the top and bottom nodes: Girder displacements
in noncomposite systems, particularly rotations, are different than girder displacements in
composite systems. To demonstrate this behavior in simple terms, an idealized deformation
pattern is provided in Figure 3-27. Figure 3-27 schematically depicts a common displaced
shape of an X-type cross-frame panel under applied vertical loads; a noncomposite system
is compared to a composite system. Note that the displacements are greatly exaggerated and
that rigid-body motion has been removed for clarity.
In comparing the noncomposite and composite systems, it is evident that noncomposite
girders generally rotate out of plane about their centroid under applied vertical loads.
Consequently, the top and bottom struts tend to deform equal magnitudes but in oppo-
site directions, which results in equal-and-opposite axial-force effects in those members.
In contrast, composite girders tend to rotate out-of-plane about an axis higher on the
cross-section due to composite action with the concrete deck. As a result, the deformation
demand on the top strut is small, as it is located near the point of girder rotation. In relative
terms, the bottom strut deforms significantly more, which is illustrated in Figure 3-27 with
the original position of the system lightly shadowing the deformed position. Under this
idealized displaced shape, the top strut sees little to no force effects whereas the bottom
strut sees substantial axial tension. To maintain compatibility with the girder displacements,
both diagonal members in the X-frame must also deform under axial tension.
Thus, when this idealized differential-rotation deformation pattern is paired with differen-
tial vertical displacement (which generates substantial diagonal forces approximately equal-
and-opposite in magnitude), the net effect is unbalanced diagonal forces. Cases in which the
diagonals are both in tension or both in compression are, therefore, not uncommon. This
generally explains why the equal-and-opposite force distribution in X-type cross-frames is
observed in noncomposite systems but not in composite systems where realistic deformation
patterns are highly complex. For K-type cross-frames, this discussion is less influential as
equal-and-opposite force distribution in diagonal members is anticipated given the layout
of the members.
Figure 3-27. Idealized deformation pattern of noncomposite (left) and composite

(right) systems.

Assuming equal shear force distribution in the postprocessing phase of 2D PEB models
generally leads to poor results for cross-frame diagonals. With that in mind, evaluating the
“true” shear force distribution from the 3D models provides an indication of how significant
the effects illustrated in Figure 3-27 are.
3. The relative difference between top strut forces and bottom strut forces: Similar to the discussion
in Item #2 above, it is anticipated that bottom strut force effects will greatly exceed top strut
force effects in composite systems. By evaluating the relative difference between these two
magnitudes, the effects of composite action can be quantified. The sign of the top strut force
can also provide an indication as to where the point of girder rotation exists along the depth
of the cross-section (e.g., if the top and bottom struts are both in tension, then the point of
rotation is likely above the top strut).
In the context of the three comparative metrics established above, the following observations
can be made from Figure 3-26:
• Total shear force: For the noncomposite system, the total cross-frame shear force estimated
from the 2D PEB model is nearly identical to that of the 3D model (i.e., 14.8 kips versus
14.6 kips). This indicates that the Timoshenko beam analogy accurately represented the
effective cross-frame stiffness by considering both shear and flexural deformations. For
reference, the shear-analogy approach produced a total cross-frame shear of 10.6 kips for this
same example, despite not being shown in the figure. This indicates that the shear-analogy
underestimates the cross-frame stiffness. For the composite system, the total cross-frame
shear force is underestimated in the 2D PEB models (e.g., 4.2 kips versus 5.7 kips). This
relationship is explored further in this section.
• Shear force distribution: For the noncomposite system, the 3D model demonstrates that
shear force is equally distributed to the top and bottom struts (subsequently resulting in
equal-and-opposite diagonal force effects). As a result, the “50-50” distribution assumption
commonly utilized in 2D postprocessing is valid for noncomposite conditions. Thus, the
2D models accurately predict the force effects in the diagonal members for the specified
loading condition.
For the composite system, the 3D analysis model demonstrates that load-induced forces
in diagonal members are not equal and opposite. Figure 3-26 shows 2.3 kips of compression
in one diagonal and 10.4 kips of tension in the other, resulting in an “80–20” distribution of
the total shear force. Thus, when comparing cross-frame forces results, it is evident that the
2D PEB postprocessing procedure produces inaccurate results.
• Moment force couple distribution: For the noncomposite system, the 3D analysis results
indicate that the load-induced forces in the top and bottom struts are nearly equal and
opposite, as expected (i.e., 21.1 kips of compression in the top strut and 20.8 kips of tension
in the bottom strut). Therefore, the 2D PEB model accurately predicts top and bottom strut
forces when compared to the 3D counterpart model.
For the composite system, this equal-and-opposite relationship does not hold true. The
3D model indicates that the top strut force effects share the same sign as the bottom strut
force effects, but the magnitude is significantly less. Based on the postprocessing assumption,
the 2D PEB model produces equal-and-opposite forces in the top and bottom struts.
Thus, there is substantial error associated with the force prediction in the top strut member
in the composite system for the 2D PEB model. However, the error in the force prediction
in the bottom strut member is minimal (i.e., within 5% conservative in both cases). Despite
the inconsistencies and oversimplifications in the 2D analysis and postprocessing, this same
general trend related to bottom strut members is typically observed assuming the equivalent
beam properties are accurate (i.e., the Timoshenko beam approach is utilized). The bottom
strut results in Figure 3-25 are additional examples of this trend.

Given that 2D PEB models generally produce reasonably accurate estimates for bottom
strut force effects, especially for simple bridge geometries, it is important to note when these
limitations and shortcomings become critical. If a bottom strut cross-frame member maximizes
load-induced force effects and governs fatigue design, then this simplified analysis and post-
processing procedure likely produces reasonable estimates of the governing design forces.
However, if diagonal members (or much less likely, top struts) govern the fatigue design,
then 2D PEB approaches may grossly underestimate the governing force effects. To expand
on the findings from Figure 3-26, which focuses on a simple straight bridge with normal
supports, the research team repeated these same studies to examine more complex bridge
geometries and loading scenarios.
Two additional bridges were examined similar to the sample bridge presented above: a straight
bridge with skewed supports (X-type cross-frames) and a curved bridge with normal supports
(X-type cross-frames). Additional studies were performed on different bridge geometries and
cross-frame configurations (i.e., K-type versus X-type); those results are presented in Appendix F.
For the straight and skewed bridge, an interior-bay cross-frame panel near the intermediate
skewed support is examined. For the curved bridge, an interior-bay cross-frame panel near
the maximum positive dead load moment region is examined.
The results outlined above investigated only two different extreme conditions: noncomposite
and full in-service composite (i.e., concrete modulus, Ec, taken as 5,000 ksi). Rather than
evaluate just the two extremes, the research team expanded the study by investigating different
variations of Ec to better quantify how composite action affects cross-frame behavior. The
concrete modulus, Ec, was varied as opposed to the deck thickness as to maintain a consistent
centroid location for the concrete deck and to simplify discussions. The Software A analysis
procedures previously described in this subsection are otherwise unchanged. As such, a total
of 272 unique models were developed as part of this study, including different bridges, various
3D and 2D modeling approaches, and multiple iterations of Ec. Note that the Ec = 0 ksi case
represents noncomposite conditions. Instead of completely removing the deck from the model,
the deck was simulated with virtually no stiffness to expedite the repetitive calculations.
The response of the specified cross-frame members in each bridge was evaluated for two
different static loading cases (100 kips in magnitude) that are arbitrarily identified as Load
Case 1 and 2. For reference, the location of load application relative to the framing plan is
sketched in Figure 3-28 and Figure 3-29. The results presented herein follow the same orga-
nization as the three comparative metrics outlined above: total cross-frame shear, shear force
distribution, and moment force couple distribution.
Figure 3-28 and Figure 3-29 show how the total shear force developed in the cross-frame
panel of interest varies as a function of the concrete deck modulus (or the degree of composite
action, in more general terms). Figure 3-28 focuses on the straight bridge with skewed supports
and X-type cross-frames, and Figure 3-29 focuses on the curved bridge. In these figures, only
the results of 2D PEB models associated with the flexural-analogy and the Timoshenko beam
approach are presented, given the inherent inaccuracies with the shear-analogy approach
outlined previously.
Figure 3-30 demonstrates how the total shear force is distributed to the top and bottom
nodes per the 3D analysis model. Since the 2D models assume a “50-50” distribution, results
related to 2D models are excluded as they do not add value to the graph. In Figure 3-31, the ratio
between the top strut force and the bottom strut force is plotted as a function of the assumed
concrete deck modulus. The results of the two sample bridges are compiled and provided. Note
that these results correspond to the 3D models only, as the 2D models assume an equal-and-
opposite force couple in the postprocessing phase. Thus, the 2D analysis results do not add
significant value to the graph.

Case 2 (100 k) VCF VCF
Case 1 (100 k)
5
Case 1 Case 2
Cross-Frame Shear, VCF (kips)
1 Model; Equivalent Beam

3D (Control)
0 2D PEB; Timoshenko Beam
2D PEB; Flexural-Analogy
-1
0 1 10 100 1000 10000 0 1 10 100 1000 10000
Concrete Modulus (ksi; log scale)
Figure 3-28. Total cross-frame shear force as a function of concrete

deck modulus (straight bridge with skewed supports and X-type
cross-frames).
The following additional notes provide important insight to the development and construc-
tion of Figure 3-28 through Figure 3-31:
• A cross-frame diagram is provided in each plot to illustrate the nodal force or internal axial
force examined in the graph. For instance, Figure 3-28 and Figure 3-29 present the definition
of the total cross-frame shear force, VCF.
• As stated above, a modulus value of 0 ksi represents a noncomposite system, whereas values
between 3,000 and 5,000 ksi are representative of most in-service, composite bridges in the
United States. The horizontal axis is presented on a log scale to clarify the noticeable change
in behavior between 100 ksi and 1,000 ksi.
VCF VCF
30
Case 1 Case 2
Cross-Frame Shear, VCF (kips)
25
20
15
10 Model; Equivalent Beam

3D (Control)
5 2D PEB; Timoshenko Beam
2D PEB; Flexural-Analogy
0
0 1 10 100 1000 10000 0 1 10 100 1000 10000
Figure 3-29. Total cross-frame shear force as a function of concrete

deck modulus (horizontally curved bridge).

Load Case 1; Vtop / VCF

Key: Vtop
Load Case 1; Vbot / VCF VCF VCF
Load Case 2; Vtop / VCF
Load Case 2; Vbot / VCF Vbot
Cross-Frame Shear Distribution, Vi / VCF

2.0
Straight, Skewed Curved, Normal
1.5
1.0
0.5
0.0
-0.5
-1.0
0 1 10 100 1000 10000 0 1 10 100 1000 10000
Figure 3-30. Shear force distribution as a function of concrete deck modulus

for two sample bridges with X-type cross-frames.
• In Figure 3-30, the shear force acting on the top node (Vtop) and the bottom node (Vbot) are
compared independently with the total cross-frame shear (VCF). Thus, the sum of all Vtop /VCF
and Vbot /VCF ratios equals one. Note that Vtop and Vbot are arbitrarily taken relative to the
left side of the cross-frame panel as shown in the sketch. Had the shear forces been taken
from the opposite side, the values of Vtop and Vbot would be flipped. The general behavior,
however, would remain the same given that the shear force is constant across the width of
the cross-frame panel.
• In Figure 3-31, the ratio of the top strut force (PTS) and the bottom strut force (PBS) is
compared to a similar ratio determined by elastic beam theory. In other words, the concrete
deck, top strut, and bottom strut are treated as three force couples that develop the moment
strength of a pseudo-composite transverse beam, similar to a reinforced concrete beam.
The elastic neutral axis of this pseudo-composite beam is derived after transforming the
PTS
Key: Elastic Beam Theory
Load Case 1 Results
Load Case 2 Results PBS
1.0
Straight, Skewed
Normal Curved, Normal
Cross-Frame Force Couple
0.5
Distribution, PTS / PBS
0.0
-0.5
-1.0
-1.5
0 1 10 100 1000 10000 0 1 10 100 1000 10000
Figure 3-31. Force couple distribution as a function of concrete deck

modulus for the two sample bridges.

effective concrete area with the corresponding modular ratio. Assuming an elastic stress
distribution, the PTS /PBS ratio is related to beam curvature and is simplified as yTS /yBS, where
yTS and yBS are the distances measured from the centroid of the respective strut to the
computed neutral axis. Therefore, by comparing the analysis results to this idealized beam
theory approach, the impact that composite action has on the distribution of bending
moments through the deck and cross-frame panel can be evaluated.
The following observations can be made with regard to the total shear force estimated by
the various analysis methods (Figure 3-28 and Figure 3-29):
• As the concrete modulus increases beyond zero (noncomposite), it is intuitive that the
cross-frame forces generally decrease. Not only does the relative stiffness of the concrete
deck increase and thus attracts more force in the highly indeterminate system, but the girder
displacements diminish as the global stiffness increases. However, as the concrete modulus
approaches values exceeding 3,000 ksi or more, the relative change is less significant.
• In general, the Timoshenko beam approach is accurate for noncomposite systems when
compared to the 3D analysis, but the variability increases slightly for composite systems.
Still, it is consistently the most accurate 2D equivalent beam method in terms of cross-frame
force predictions, as is presented later in this subsection.
• The observed behavior for Load Case 2 in Figure 3-28 is less predictable. Note that the vertical
axis scale is much smaller than in Figure 3-29 due to the nature of the loading. In Figure 3-29,
the applied load is in close proximity to the cross-frame panel of interest, such that the shear
demands are more significant. In Figure 3-28, though, the applied loads induce a torsional
response to the superstructure, such that the net shear on the cross-frame of interest is
generally small. Consequently, the discrepancy in the total cross-frame shear is less influential
in terms of predicted cross-frame force effects, as is presented later.
The following observations can be made regarding the shear force distribution determined
by 3D analyses (Figure 3-30):
• For noncomposite systems (i.e., Ec = 0 ksi), especially for the non-skewed bridge, the shear
force is equally distributed top and bottom (same result as Figure 3-26 before). The distribu-
tion for the straight bridge with skewed supports is slightly less balanced given the nature of
the cross-frame deformation pattern.
• As the concrete deck becomes stiffer and composite action becomes more substantial, it is
evident that the shear force distribution gradually diverges from the “50-50” assumption. For
instance, the top shear reaction in the straight, skewed bridge for Load Case 2 is approximately
1.6 VCF, whereas the bottom shear reaction is larger than 0.6 VCF in the opposite sign. In this
particular instance, the observed behavior indicates that both diagonals were in significant
compression.
The following observations can be made regarding the force couple distribution determined
by 3D analyses (Figure 3-31):
• For the noncomposite systems, the PTS /PBS ratio consistently converges to -1, which represents
an equal-and-opposite force distribution in the struts. As the deck stiffens and composite
action become more significant, the ratio tends to a small positive value. This indicates that
the top strut shares the same sign as the bottom strut, but the magnitude is substantially
smaller (i.e., a clear indication of composite behavior). In fact, the transition between a nega-
tive ratio and a positive ratio as Ec increases represents the condition in which the center of
out-of-plane girder rotation moves above the top strut. At that point, the top and bottom
struts would be both compressive or both tensile forces.
• In general, the analysis results follow the same trends as the “elastic beam theory” bench-
mark solution, but not exactly. Thus, the flexural action assumed in the development of this

idealized curve is apparent in the realistic data, but the deviation from that curve is due to the
fact that the “true” deformation patterns are not identical to the idealized one.
With general observations established about the validity of common analysis and postpro-
cessing procedures, the next step is to compare the predicted force effects between 3D and 2D
models. Utilizing those common assumptions (i.e., equivalent beam moment resolved as a force
couple and equivalent beam shear equally distributed to top and bottom nodes of X-frames),
2D PEB results are compared to the 3D solutions. The results of the two sample bridges are
compiled and presented in Figure 3-32. In total, four different graphs are presented. The two
columns represent the two distinct load cases examined for each bridge. The two rows represent
the different bridges evaluated as part of this study.
The horizontal axis represents the elastic modulus of the concrete deck and is presented on
log scale. The vertical axis represents the relative error of the 2D PEB model relative to the
corresponding 3D model. Note that only the results using the Timoshenko beam analogy are
shown. Relative error values of zero indicate perfect agreement between the 3D model and
the simplified 2D model. Relative error values exceeding 1.0 indicate the 2D models over
estimate the force effects in excess of 100%; the opposite is true for relative errors below –1.0.
For clarity, data points beyond -1.0 and 1.0 are eliminated from view.
The relative errors associated with the diagonal members and the bottom strut member
are also graphed separately. The lines are color-coordinated based on the cross-frame sketch
provided in the corners of each graph. The results for top struts are excluded given that they
are generally lightly loaded members.
The following observations can be made regarding the relative error associated with the
simplified 2D modeling procedures (Figure 3-32):
• For noncomposite bridges, relative error approaches zero for all member types and all bridge
types. Although not explicitly shown in the report, this is especially true for straight and
Case 1 Case 2
1.0 Key:
0.5
Straight, Skewed
Relative Error (Compared to 3D Model)
0.0
Key:
-0.5
-1.0
1.0 0.1 1 10 100 1000 0.1 1 100 1000 10000
0.5
Curved, Normal
0.0
Key: Key:
-0.5
-1.0
0 1 10 100 1000 10000 0 1 10 100 1000 10000
Figure 3-32. Analysis error of the 2D PEB model as a function of concrete

deck modulus for the two sample bridges.

normal bridges. This is largely attributed to two ideas: (i) the Timoshenko beam approach
accurately represents the cross-frame stiffness in the analysis, and (ii) the common post
processing procedures are valid for noncomposite systems. The error increases slightly with
more complex bridge systems (i.e., skewed or curved bridges) but is still relatively accurate.
For example, the relative error associated with the “red” X-frame diagonal in the skewed
bridge for Ec = 0 ksi is approximately 20% conservative under Load Case 1. Under Load
Case 2, the relative error is not shown largely due to the fact the absolute magnitude of the
force is small. In this case, any slight discrepancies typically result in large error values.
• For the composite systems evaluated (i.e., Ec exceeding 3,000 ksi), the error associated with
bottom strut members is typically small (within 5%–10%). Despite its limitations and
oversimplifications, 2D methods produce accurate force estimates for bottom strut members.
This is consistent with the results presented previously in this section.
• For diagonal members in X-type cross-frames, the errors generally become excessive as Ec
increases. This trend was also observed for straight bridges with normal supports. One diagonal
tends to a large positive value, and the other tends to a large negative value. For example, the
relative error associated with the diagonal members of the curved bridge under Load Case 2
approach +100% and -100% error, respectively, for Ec = 5,000 ksi. This is attributed mainly
to the equal shear force distribution assumed with the postprocessing methods.
• Although not explicitly shown in the report, the errors for diagonals in K-type cross-frames
are much more manageable than those with X-type cross-frames. For these cross-frame types,
recall that the distribution of shear force distribution is not a meaningful discussion.
• Although not explicitly shown, relative error magnitudes associated with 2D PEB models
using the flexural-analogy approach for equivalent beams are more substantial than those
shown in Figure 3-32 for the Timoshenko beam approach. Thus, it can be concluded that the
Timoshenko beam approach is more appropriate for noncomposite and composite systems.
From Figure 3-32, it is apparent that 2D analysis methods produce poor cross-frame force
estimates in two areas: (i) in top strut members and (ii) in X-frame diagonal members regardless
of bridge type. Erroneous, overly conservative results in top struts are not generally impactful
in terms of cross-frame fatigue design. These members typically do not govern fatigue design
such that those results, accurate or not, can be disregarded. Erroneous diagonal member forces,
on the other hand, can have significant impact on the cross-frame fatigue design of a bridge.
As such, it is prudent to reevaluate how designers typically handle the postprocessing of 2D
analysis models to ensure more reliable results.
The major challenge in developing a simple postprocessing modification is the degree in
which the shear force distribution varies. In noncomposite systems, diagonal forces typically
are nearly equal-and-opposite. In the composite systems evaluated as part of Figure 3-30, shear
force distribution was shown to vary from nearly equal to 150% VCF at one node and –50% at
the other. The distribution is a function of bridge type, loading, and subsequently the induced
deformation pattern, which is difficult to quantify without developing a 3D model. An alter-
native method is explored in Section 3.4.4.
3.4.3 Parametric Study

To further assess the performance of 2D analysis methods, a 20-bridge parametric study
(detailed in Section 2.5.4) was conducted. For each cross-frame member evaluated, the maxi-
mum force range was computed based on the critical lane passage of the AASHTO fatigue truck.
Thus, the results in this section differ slightly than the previous results in Section 3.4.2, which
were based on static loads.
It should be noted that, in some cases, the governing lane conditions between the 3D and
simplified 2D models differed. For example, a 3D model indicated that a specific cross-frame

member under a specific transverse lane passage maximized force effects, whereas the cor
responding 2D model indicated that the force in the same member was maximized by a
different lane passage. To maintain consistency, the governing lane position was established
based on the control model (i.e., the 3D validated model in Abaqus); the results of the simplified
analysis models were then obtained from those established parameters.
Figure 3-33 presents this data in the form of a box-and-whiskers graph, similar to those pre-
viously shown in the report. Each box-and-whiskers plot represents a set of unfactored design
force effects obtained from the various analysis models. Along the vertical y-axis, the governing
force effects derived for each iteration of the 20 representative bridges are compared to the
control model. The percent error with respect to the results of the control model is reported.
Zero percent error indicates perfect agreement between the simplified commercial design software
models and the control model. Positive error indicates that the simplified analysis conservatively
overestimates the predicted force effects, and negative error indicates the opposite.
In terms of accuracy, it is desirable that the simplified methods provide solutions within about
10% of the validated solution with marginal variance in the box-and-whisker. In terms of a
conservative design approach, it is preferred to have positive error over negative error.
In Figure 3-33, the results corresponding to bottom strut members are differentiated from
diagonal members based on the discussion in Section 3.4.2. In both plots, the results are catego
rized by the modeling technique along the horizontal x-axis (3D; 2D grillage with flexural-
analogy equivalent beams; and 2D PEB with shear-analogy, flexural-analogy, and Timoshenko
equivalent beams).
From Figure 3-33, the following generalized observations can be made when comparing the
analysis error reported across the full data set:
• The 3D commercial design software analysis generally provides cross-frame results (struts
and diagonals) that agree well with the control model. On average, the predicted design force
range was 2% conservative. The observed scatter in the results was also small.
• For the 2D analysis methods, significant variability is observed. This behavior is attributed to:
(i) skewed and/or curved geometry and (ii) the postprocessing errors inherent with the 2D
analysis methods outlined in Section 3.4.2.
100% Bottom Struts Diagonals

80%
60%
40%
Percent Error
20%
0%
–20%
–40%
–60%
–80%
–100%
Figure 3-33. Box-and-whiskers plot demonstrating the range of

analysis error obtained from the Commercial Design Software Study
(bottom struts versus diagonals).

• The 2D PEB models (shear-analogy) consistently underpredict the governing design force
range (-80% error on average). The 2D PEB models (flexural-analogy) improve upon the
shear-analogy approach but still underpredict the design force ranges.
• The 2D PEB models (Timoshenko beams) provide the most accurate solutions with respect
to 2D models but still result in significant errors. These models are generally conservative
(between 0% and 20% conservative) for bottom struts but are unconservative for diagonals
(between 40% and 60%). This is attributed to the equal shear assumption used in the post-
processing procedure.
• 2D grillage models using the flexural-analogy for the development of equivalent cross-
frame beams show considerable scatter in the predicted cross-frame response. In some cases,
the model predicted design forces above 100% conservative and below 60% unconservative.
The upper tail of the bottom strut box-and-whiskers results is beyond the scale of the vertical
axis, given the large discrepancies observed.
Figure 3-34 expands on Figure 3-33 by breaking up the bottom strut results based on bridge
geometry. For the 2D PEB results, only models utilizing the Timoshenko beam approach are
included for clarity. Note that the skew index and connectivity index established in NCHRP
Research Report 725 (White et al. 2012) and defined in Eqs. 3.1 and 3.2 are used to differentiate
straight and curved bridges with or without support skews.
To explore the effects of bridge geometry, Figure 3-34 compares the results in terms of skew
and connectivity indices. Recall that only bottom strut results are presented in this figure for
clarity. The following generalized observations can be made from that figure with an emphasis
on PEB methods:
• 3D models generally perform well for all cases. Thus, 3D models offer consistently reliable
results in terms of cross-frame forces primarily because the superstructure depth is accurately
represented.
• 2D PEB models are generally conservative in terms of their bottom strut force estimates.
However, scatter in the results is more evident as the bridges become more skewed
and/or more curved. For instance, the percent errors range from –20% unconservative to
20% conservative for straight and normal bridges, whereas the percent error ranges from
–60% conservative to 80% conservative for curved and skewed bridges.
• Although not presented, the results for diagonal members are more variable than what is
shown in Figure 3-34 for bottom struts.
100%
80%
60%
40%
Percent Error
20%
0% 2D
Grid
-20% 3D 2D 3D 3D
-40% PEB 2D
2D 3D
PEB
-60% 2D PEB 2D 2D
2D
-80% Grid Grid PEB
Grid
-100%
Straight & Normal Curved & Normal Straight & Skewed Curved & Skewed
IS ≤ 0.1 IS = 0 IS > 0.1 IS > 0.1
IC ≤ 0.5 IC > 0.5 IC = 0 IC > 0.5
Figure 3-34. Box-and-whiskers plot demonstrating the range of analysis

error obtained for different bridge geometries (bottom struts only).

It is evident from the results that 3D models offer increased repeatability and reliability
in terms of cross-frame force effects. However, the biggest setback to simplified 2D analyses
(particularly PEB models) is related to diagonal members and the postprocessing assumptions.
This statement is independent of bridge type, although analysis error becomes more apparent
for complex systems. Consequently, Section 3.4.4 explores a simplified method to enhance
the postprocessing procedures for 2D analysis methods.
3.4.4 Alternative 2D Postprocessing Procedure

The research team investigated potential methods of manipulating 2D grillage and PEB
analysis results to produce more reasonable estimates of cross-frame design forces. As discussed
previously, equivalent beam moments from grillage models (excluding those that use notional
beams to represent the transverse stiffness of the deck) can be resolved based on the equivalent
truss/deck model illustrated in the left side of Figure 2-17. Equivalent beam models from PEB
models, despite providing equal-and-opposite strut force effects, have been shown to produce
reasonable estimates of the more critical bottom strut members. However, the distribution of
the shear force in both grillage and PEB models directly impacts the diagonal member force
effects, which are highly variable.
The research team explored procedures to resolve this issue. A simple alternative to the
“50-50” distribution is to conservatively design each diagonal member for 100% of the total
cross-frame shear. That is to say, the top and bottom struts are handled in the same manner
as presented above (i.e., equal-and-opposite force distribution, but neglect the top struts in
design), but the diagonals are handled differently. This procedure is schematically shown in
Figure 3-35, which is a modified adaptation of Figure 3-26.
First, note that the 2D analysis output (i.e., end shear and moments on the equivalent cross-
frame beam) is identical to that of Figure 3-26. Second, the top and bottom strut forces are
computed (i) assuming the end moment of 756 kip-in is applied as a force couple to the top and
bottom nodes and (ii) assuming the end shear of 4.2 kips is equally distributed top and bottom.
This produces a top strut design force equal to 8.5 kips of compression and a bottom strut design
force equal to 8.5 kips in tension, which is the same as Figure 3-26. In this instance, the top strut
force can be disregarded, and the bottom strut force from the 2D analysis (8.5 kips) is in good
agreement with the 3D analysis (8.4 kips).
Lastly, the diagonal forces are calculated independently assuming 100% of VCF is applied
to the node of interest. In Figure 3-35, the full 4.2 kips of shear are applied to the bottom node
to maximize force in the tension diagonal (9.4 kips), which shows much better agreement to the
3D analysis (10.4 kips) than was presented in Figure 3-26 (4.6 kips). A similar procedure for
the top node was performed, although not explicitly shown in the figure, which results in
9.4 kips of compression in the opposite diagonal. This simple method, despite producing
equal-and-opposite diagonal forces, produces design magnitudes generally closer to the 3D
analysis (at least for the critical diagonal member).
It is acknowledged that there are obvious limitations to this simplified method. Given that
the true governing member of this cross-frame panel would be the tension diagonal, designing
for 9.4 kips of compression is overly conservative. It must also be recognized that this procedure
could potentially be overly conservative in some cases, where the true shear distribution is closer
to “50-50” (e.g., Load Case 1 in curved bridge or noncomposite systems in general). Similarly,
this procedure could still be unconservative for cases in which the shear at one node exceeds
100% VCF (e.g., Load Case 2 in a straight, skewed bridge). As stated above, this simple procedure
is one of many possible solutions to this problem. Yet, the only way to ensure reliably accurate
results, regardless of bridge type, is to develop a 3D model.

100 k
2D PEB (Composite System) 3D (Composite System)
Analysis Output:
756 k-in 258 k-in
4.2 k 4.2 k
Postprocessing Strut Forces:

12.6 k 8.5 k 4.3 k
2.1 k 2.1 k
60”
2.1 k 2.1 k
8.5 k 2.6 k
12.6 k 4.3 k
120”
Postprocessing Diagonal Forces:

8.4 k
4.2 k
4.2 k
Combining Results:
8.5 k
8.5 k
Figure 3-35. Alternative postprocessing procedure demonstrated

through Load Case 1 on sample straight bridge with normal supports
(work with Figure 3-36).

Based on the results of the Commercial Design Software Study presented, there are several
conclusions that can be drawn with respect to simplified analysis methods for cross-frame design:
• In general terms, designers must be cognizant of the potential trade-offs between sophisti-
cated and simplified analysis (in terms of accuracy and ease of use). A central theme explored
throughout the report is related to the balance between increased computational complexity
and improved reliability versus simplified modeling and improved ease of use. The results
presented in this section clearly highlight this trade-off. The designer must make informed
decisions in terms of the analysis performed. The conclusions below are intended to provide
practicing engineers the technical and quantitative background to help inform those decisions,
while not necessarily advocating one or the other for a specific project need.
• In all cases, 3D refined analysis produces improved accuracy for predicting cross-frame forces
compared to 2D analytical approaches. Although 3D models are more time-consuming to

develop than the 2D counterparts, these models offer solutions with improved accuracy,
reliability, and repeatability. The traditional approach of connecting cross-frame members
into a shared node along the web-to-flange juncture is also acceptable.
• 2D PEB models generally produce more accurate results than 2D grillage models because the
concrete deck is explicitly considered. Although there are methods to consider the effective
transverse stiffness of the deck, it is more accurate to represent the deck as a thin shell element
that is rigidly connected to the girder to simulate composite action.
• Common postprocessing practices for 2D methods tend to produce erroneous results for
top strut members and inaccurate results for diagonal members in X-frames, regardless of
bridge type. A simple, conservative approach has been presented, but alternative methods
can and should be explored by designers or programmers.
• If a designer elects to use a 2D modeling approach for these complex cases, it is recommended
to incorporate simple modifications to the model (e.g., utilizing the Timoshenko approach
for equivalent cross-frame beams and considering the transverse stiffness of the concrete for
grillage-type models) and postprocessing (e.g., conservatively applying the full cross-frame
shear force in the design of X-frame diagonals and considering the concrete deck in post-
processing of grillage models similar to Figure 2-17). Still, even after implementing these
improvements to 2D analyses, substantial error is likely in cross-frame force predictions,
especially in bridges with significant horizontal curvature and/or support skews.
• Based on the conclusions above, it is recommended that any bridge satisfying the geometric
limits established in Section 3.2.4 (i.e., bridge geometries for which live load force effects in
cross-frames should be considered) should be evaluated using 3D modeling techniques. In
general terms, this implies that heavily skewed and/or curved systems are best suited to be
analyzed with 3D models.
and commentary.
3.5 Stability Study

This section summarizes the results related to the Stability Study, which focused on Objective
(e) of NCHRP Project 12-113. More specifically, the appropriate stability bracing strength and
stiffness requirements for implementation into AASHTO LRFD are examined computationally.
These goals were summarized by the following questions that were initially posed in Section 1.2:
• Can the AISC design guidelines for stability bracing be incorporated into AASHTO LRFD?
• Are special requirements needed for negative moment regions of continuous systems?
• How are these stability bracing requirements combined with other load conditions such
as wind?
These questions are systematically addressed throughout this section of the report. Before pre-
senting the results of the bracing studies, Section 3.5.1 documents a computational study on the
buckling behavior of girders in the composite condition. Section 3.5.2 presents the key results
of the bracing strength study, and Section 3.5.3 outlines the key results of the bracing stiffness
study as well as modifications to the AISC Specification expressions based on the major find-
ings. Section 3.5.4 provides an overview of how to implement these cross-frame stability bracing
forces in the context of a design. Lastly, Section 3.5.5 summarizes the outcomes of these studies.
3.5.1 Buckling in the Composite Condition

As noted in Section 2.6.1, an elastic eigenvalue buckling analysis was carried out to examine
the effects of continuous top flange restraint on the buckling behavior of bridge girders. In this

section, several sample results are presented that demonstrate the benefits of a composite con-
crete deck on the buckling capacity of the steel girder section, even in the region around interior
supports when shear studs may not be provided and the moments cause compression in the
bottom flange.
Figure 3-36 presents sample eigenvalue buckling results of a girder with an unbraced-length-
to-depth ratio (Lb/d) of 5 and a well-stiffened web with various top flange restraints and moment
gradients. For the three different top flange restraints considered (i.e., no additional restraint,
continuous lateral restraint, and continuous lateral and torsional restraint), the critical buckling
moment, Mcr, was obtained from the FEA analysis and is plotted. To demonstrate the magni-
tude of these buckling loads compared to a limit state governed by cross-sectional yielding,
Mcr is normalized by the moment at first yield of the section, My, where the yield strength
is taken as 50 ksi. Along the horizontal axis, the results are organized by the corresponding
straight-line moment gradient, which is represented as the ratio between applied end moments,
MR/ML (applied moment at the right end of the unbraced beam segment/applied moment
at the left end). For example, MR/ML = 1.0 represents a uniform-moment case in which the
bottom flange is subjected to uniform compressive stresses. Sketches are provided above the
figure to graphically illustrate these moment gradients.
From Figure 3-36, the following observations can be made with regards to the effects of top
flange restraint on girder buckling behavior:
• In general, the critical buckling moment increases as MR/ML decreases due to the benefits of
moment gradient along an unbraced segment in terms of buckling capacity. The intent of
the moment gradient factor, Cb, adopted by many design specifications is to account for
these effects in design.
• In comparing the case with no additional top flange restraint and the case with continuous
lateral restraint, it is evident that adding continuous lateral support increases the buckling
capacity, particularly for cases with reverse-curvature bending (i.e., MR/ML = -1). These results
are consistent with the modified Cb factor expressions in the AISC Commentary Section F1
(2016) that account for continuous lateral restraint.
• When adding continuous lateral and torsional restraint, the critical buckling moments
consistently exceeded the yield moment of the section. For these cases, the buckled shapes
of the beams were consistent with the web distortion mode presented in Figure 2-18. Even
for the uniform-moment condition in which the restrained top flange is purely and uniformly
ML MR ML ML MR
Normalized Buckling Moment, Mcr / My
3.0
Key: Top Flange Restraint
2.5
Continuous Lateral &
2.0 Torsional
1.5
Continuous Lateral
1.0
0.5
None
0.0
–1 0 1
Applied End Moment Ratio, MR/ML
Figure 3-36. Normalized critical buckling moment as a function of

moment gradient and top flange restraint (Lb/d = 5; stiffened web).

in tension, Mcr exceeded My by approximately 50%. Thus, these unbraced segments would
yield prior to reaching an instability, which in turn mitigates the bracing demands on the
cross-frames in these regions.
To expand on Figure 3-36, Figure 3-37 evaluates two different conditions independently:
an unbraced-length-to-depth ratio of 10 and an unstiffened web. The intent of this figure is
to examine the effects of increased unbraced length and web slenderness on the LTB and web
distortion buckling modes introduced schematically in Figure 2-18. Recall that the cases for
which no top flange restraints are provided (blue line in Figure 3-37) and continuous lateral
restraints are provided (red line) generally correspond to more conventional lateral-torsional
buckled shapes. In contrast, the case simulating composite systems (i.e., continuous lateral
and torsional restraint to the top flange) produced buckled shapes more consistent with web
distortion modes.
From Figure 3-37, the following observations can be made with regards to the effects of
unbraced length and transverse stiffeners on LTB and web distortion buckling:
• As demonstrated by the left figure, increasing the unbraced length impacts the LTB modes
(i.e., girders with no top flange restraint and girders with continuous lateral top flange
restraint) much more significantly than the web distortion modes (i.e., girders with con-
tinuous lateral and torsional top flange restraint). For instance, a 70% reduction in buckling
capacity was consistently observed when the Lb /d ratio was doubled. These results verify the
findings in Helwig and Yura (2015) that state that web distortion buckling is not as sensitive
to the unbraced length.
• As demonstrated by the right figure, eliminating the web stiffeners drastically reduces the
buckling capacity of the web distortion buckling modes. A nearly 70% decrease in Mcr on
average was observed when compared to the finite element solutions for the girders with
well-stiffened webs. For the LTB-related cases, this trend is less pronounced, particularly
for the uniform-moment case where no shear force is present. For the cases with moment
gradient, however, a slight reduction in Mcr is observed due to the effects of web distortion
on the LTB response, which were intrinsically neglected in the development of the tradi-
tional LTB solutions developed by Timoshenko and Gere (1961). Note that the solution
derived by Timoshenko assumes a perfectly straight web (i.e., absent of any distortion) in
its buckled shape. Studies have shown that the interaction between web distortion and LTB
is most pronounced with smaller unbraced lengths, which are often controlled by yielding
over buckling.
Key: Top Flange Restraint

20%
Stiffened Web Unstiffened Web
Continuous Lateral & Lb /d = 10 Lb /d = 5
0% Torsional
Reduction in Mcr
-20%
-40%
Continuous Lateral
-60%
None
-80%
-1 0 1 -1 0 1
Applied End Moment Ratio, MR/ML
Figure 3-37. Reduction in the critical buckling moment (compared

to the results in Figure 3-36) for (left) a longer unbraced length and
(right) an unstiffened web.

• Despite the nearly 70% reduction demonstrated in the right figure for the continuous lateral
and torsional restraint condition, it is important to note that these results are presented
in relative terms. In absolute terms, a 70% reduction in M cr for this restraint condition
still produces buckling capacities that exceed the lateral-only or no top flange restraint
conditions.
With these factors in mind, it is evident that instability related to conventional LTB behavior
is generally not critical once the composite deck has cured, and it is able to provide continuous
lateral and torsional restraint to the girder top flange. In these instances, the critical unbraced
segment in negative flexure is generally controlled by yielding of the cross-section, for which
stability bracing demands in the cross-frames are greatly diminished.
While including minimal shear studs in the negative moment region improves the twist
restraint of the girders, the deck still provides considerable restraint even without shear con-
nectors. The restraint comes in the form of “tipping restraint,” in which twist of the girder
shifts the contact point between the deck and the girder out to the edge of the flange. Tipping
restraint is very significant in most practical problems such that conventional LTB in the
composite girder is not generally a critical mode in the finished bridge. As such, the subsequent
sections focus primarily on bracing requirements for the noncomposite system, where stability
is most critical.
3.5.2 Bracing Strength Study

As outlined in Section 2.6.2, a finite element parametric study was developed and conducted
to assess two different variations of the AISC cross-frame bracing strength requirements (i.e.,
Eq. 2.3 versus Eq. 2.4). A companion study was carried out on the stability brace strength
requirements as documented in Liu and Helwig (2020). The results of these computational,
second-order studies are briefly summarized herein. The relationship between internal girder
moment and brace moment is specifically examined for a variety of bracing configurations,
girder cross-sections, and loading conditions. For clarity, only select results are presented
herein to highlight key trends that support the final suggestions and conclusions outlined in
Chapter 4. For a full overview of the results, refer to Appendix F.
In general, moment demands in the critical bracing elements were monitored as a function
of increasing girder moments. That is to say, the out-of-plane girder deformations and cor
responding brace moments resulting from second-order effects on the initial imperfection were
recorded during these incremental analyses. Thus, the figures presented in this section graph
the brace moment, Mb, as a function of the maximum girder moment, M.
As an example, Figure 3-38 illustrates this relationship for two different uniformly distributed
loading cases. Uniform-moment conditions were also evaluated as part of this study. However,
it was observed that uniformly distributed loads that introduce shear on the cross-section
produced larger moment demands on the critical braces. As such, uniform-moment results
are included in Appendix F for reference but are not included in the report.
The graphs on the left represent top flange loading, whereas the graphs on the right represent
mid-height loading. The figure also normalizes girder moments M to Mcr, which represents the
theoretical Timoshenko and Gere (1961) solution for critical elastic buckling moment between
brace points. Additionally, the brace moments, Mb, are normalized to the girder moments, M,
such that a direct comparison to Eq. 2.3 and Eq. 2.4 can be made. In other words, the cor-
responding Mb /M ratio as the applied moment, M, approaches Mcr (i.e., M/Mcr → 1.0) is the
primary focus of the FEA results.

1.2 n = 1, top flange loading n = 1, mid-height loading
1.0
0.8
0.6
0.4 Cross-section 1 Cross-section 1

Cross-section 2 Cross-section 2
0.0
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.00 0.02 0.04 0.06 0.08 0.10 0.12
1.2
n = 3, top flange loading n = 3, mid-height loading
1.0
0.8
M / Mcr
0.6

0.0
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.00 0.01 0.02 0.03 0.04 0.05 0.06
1.2 n = 5, top flange loading n = 5, mid-height loading

1.0
0.8
0.6

0.0
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.00 0.01 0.02 0.03 0.04 0.05 0.06
Mb / M
Figure 3-38. Brace moments as a function of applied bending moment

(under transverse loading).
Figure 3-38 also represents the results related to different girder cross-sections and bracing
schemes. Girder Cross-sections 1, 2, and 3 are evaluated independently as denoted on the line
graphs. In terms of bracing, note that only an odd number of intermediate braces (i.e., 1, 3,
and 5 intermediate braces) between the end supports is presented, as these cases align a brace
point at the location of maximum girder moment (i.e., at midspan). Cases with an even number
of braces were also analyzed, but those conditions resulted in less severe brace forces.
As noted in Section 2.6.2, it was observed that elastic materials (which were controlled by
stability limit states and not yielding) always resulted in larger buckling-related deformations
and thus larger brace moments compared to brace moments that occurred at the limit state of
yielding in the girder. As such, Figure 3-38 and subsequent figures presented in this section are
based on elastic material properties.

From Figure 3-38, the following observations can be made with regards to brace strength
requirements and uniformly distributed loading:
• For these single-curvature bending cases, the brace moments ranged from 3% to 10% of
the girder moment depending on the position of the load on the cross-section (load-height
effects) and the specific cross-section. As such, the current 15th Edition AISC (2016) brace
strength equation (Eq. 2.4) that recommends 2% of the design moment is unconservative
(i.e., 3% versus 2% is a 100% error).
• Top flange loading is generally more critical than mid-height loading, and girders with fewer
intermediate braces result in larger normalized brace moments. In other words, the relative
magnitude of the brace forces at the location of the maximum girder moment decreases with
an increased number of intermediate braces.
• For most analyses, slender sections (i.e., smaller bf /d ratios) resulted in larger brace forces;
however, in some cases the stockiest section (Cross-section 3) resulted in the largest brace
force as a percentage of the applied moment. The much wider flange for Cross-section 3 leads
to a larger critical brace moment compared to Cross-sections 1 and 2, such that the applied
moment M is relatively large.
• Some of the analyses did not reach convergence beyond 90%–98% of Mcr due to excessive
girder deflections. However, the original study focused on the current AISC stiffness recom-
mendation. As noted later, the recommended stiffness for the AASHTO LRFD provisions is
higher, which will result in better control (and analysis convergence) of brace forces.
As Figure 3-38 presented sample results from the parametric study, Table 3-12 summarizes
the torsional bracing strength requirements for the entire data set. As noted above, different
cross-sections, numbers of intermediate braces, and loading conditions were evaluated. The pre-
vious figure demonstrated the curves tend to flatten as M/Mcr approaches 1.0, and convergence
was not reached above 90%–98% of Mcr in some instances. In Table 3-12, the brace moments
are presented for either M/Mcr = 1 or the largest load that convergence was achieved, which was
typically capped at M/Mcr = 0.9.
Although some of the brace moments tabulated are approximately 2% at M/Mcr = 0.9 (for
analyses that did not converge), it should be emphasized that brace moments tend to increase
dramatically as the applied moment increases from M/Mcr = 0.9 to M/Mcr = 1. The brace moment
has been shown to grow between 50% and 100% beyond an applied moment of 90% of Mcr.
As such, these cases, which may not appear critical, can be rather significant.
Another point to note is that the results documented in this section were from a com-
panion study to NCHRP Project 12-113 that is documented in Liu and Helwig (2020). A later
companion study that is discussed in the next subsection modified the stiffness requirements.
Table 3-12. Summary of strength requirements of torsional bracing for different

cross-sections and load conditions.
Loading n 1 3 5
Conditions bf / d 1/6 1/4 1/3 1/6 1/4 1/3 1/6 1/4 1/3
Uniform Mb / M 6.9% 3.9% 3.7% 2.7% 2.2% 2.2% 2.0% 2.0% 3.0%
Moment M / Mcr 100% 100% 100% 90% 90% 90% 90% 90% 90%
Top Flange Mb / M 10.4% 7.3% 7.7% 2.5% 2.0% 2.0% 4.0% 3.4% 2.0%
Loading M / Mcr 100% 100% 90% 90% 90% 90% 100% 100% 90%
Mid-Height Mb / M 7.3% 4.0% 3.3% 2.5% 2.0% 2.0% 4.0% 3.4% 2.0%
Loading M / Mcr 100% 100% 100% 90% 90% 90% 100% 100% 90%

The larger stiffness requirements produce much better agreement of the strength behavior
with regards to Eq. 2.3.
In general, the brace moments reported are consistently much larger than the 0.02M
requirement given by the current AISC expression (Eq. 2.4). In many cases, the FEA results
were more than 100% larger. Therefore, it is recommended that designers use the brace strength
equation from the 14th Edition Specification with the modifications outlined in Section 3.5.3.
Even with the potential implementation of Eq. 2.3 in AASHTO LRFD, it is evident from
Table 3-12 that the required brace moments can still exceed the brace moment given by Eq. 2.3
(i.e., a minimum brace moment of 2.6% of Mcr). However, there are mitigating factors that
lessen the demands reported in the table. As previously noted, stability brace moments are
extremely sensitive to the shape and distribution of the initial imperfection in the girders.
In most situations, though, the actual imperfection will not match the critical shape considered
in this study, resulting in smaller brace forces. Also, in many situations, the brace sizes often
utilized in design will exceed the stiffness required by Eq. 2.1, which will typically result in
smaller stability-induced forces. Lastly, a companion study related to bracing stiffness require-
ments that is documented in Section 3.5.3 proposes an increase of the stiffness requirements
such that better agreement is achieved with Eq. 2.3. With that in mind, Eq. 2.3, as modified in
the next subsection, is deemed suitable for implementation into AASHTO LRFD.
Given that Table 3-12 focuses entirely on single-curvature bending cases, it is prudent to
verify these conclusions for reverse-curvature bending. To demonstrate bracing demands in
beams subjected to significant moment gradient, sample results are presented in Figure 3-39.
[Note: Mcr based on largest moment]

1.20
1.00
0.80
Key: Key:
0.60
M M
0.40
0.20
Imperfect TF Imperfect TF
0.00
M / Mcr
1.20
1.00
0.80
Key: Key:
0.60
0.40
0.20
Imperfect TF Imperfect BF
0.00
0.00 0.01 0.02 0.03 0.04 0.05 0.00 0.01 0.02 0.03 0.04 0.05
Mb / M
Figure 3-39. Brace moments as a function of reverse-curvature

bending (Cross-section 1; n = 5; mid-height loading) (TF = top flange;
BF = bottom flange).

Beams with Cross-section 1 properties were subjected to the reverse-curvature moment gradient
outlined in Section 2.6. Five intermediate braces were incorporated to maximize the negative
moment magnitude at the first intermediate cross-frame line. For the sake of comparison,
Figure 3-39 presents the results for four different loading and critical imperfection scenarios:
• Uniform moment with the critical imperfection along the top flange at midspan;
• Single-curvature, uniformly distributed mid-height loading with the critical imperfection
along the top flange at midspan;
• Reverse-curvature, uniformly distributed mid-height loading with the critical imperfection
along the top flange at midspan; and
• Reverse-curvature, uniformly distributed mid-height loading with the critical imperfection
along the bottom flange at the first intermediate cross-frame line.
Note that in the third and fourth scenarios listed, the loading conditions were identical.
Only the location of the critical imperfection was varied to study its impact on the results.
In Figure 3-38, only the midspan cross-frame results were provided, as bracing demands
were maximized at this location for single-curvature bending (due to girder moment magni-
tudes and girder displacements). For reverse-curvature bending, however, that behavior is
less clear. As such, Figure 3-39 presents the bracing force demands on three different cross-
frames in the five-brace scheme. It should be noted that the results are not symmetric (e.g.,
cross-frame lines 1 and 5 produce different results) due to the asymmetric nature of the
initial imperfection. Thus, only the results related to the more critical half of the beam are
presented.
Figure 3-39 is otherwise constructed similarly to Figure 3-38 with one key exception. Along
the horizontal and vertical axes, the maximum applied moment, M, is taken as the maximum
negative moment at the support for consistency. Mcr is then based on the end segment, which
was shown to buckle before the segments in the positive moment region by eigenvalue analysis.
This is clearly denoted at the top of the figure. Sketches are also provided for each loading and
imperfection condition, for reference.
The following observations can be made from the results presented in Figure 3-39:
• In the uniform-moment case, it is evident that the midspan brace experiences the largest
bracing demands. This validates the assumption used throughout the development of the
single-curvature cases (e.g., Figure 3-38). Additionally, the midspan brace moment approaches
2% of the maximum applied moment at M/Mcr = 0.9 (i.e., analysis did not converge at 100%
of Mcr). This result is consistent with Table 3-12.
• In the single-curvature uniformly distributed load case, it is also evident that the midspan
brace experiences the largest bracing demands. Note that the critical values are also consistent
with what is reported in Table 3-12.
• For the reverse-curvature bending cases, the critical cross-frame coincides with the location
of the critical imperfection, as expected. In the bottom left plot (imperfection along the top
flange at midspan), the bracing moment approaches 1% of the maximum applied moment
(taken as the negative moment value at the support) as Mcr is reached. The other two braces
that are presented pick up nearly no bracing forces, even at the elevated load magnitudes.
In the bottom right plot (imperfection along the bottom flange at the first brace line), the
critical bracing moment approaches 1.5% of the maximum applied moment. These cases
still produce less force demands than the comparable single-curvature conditions. Note that
the beams at the supports in these analyses were assumed to have zero initial twist. This is
consistent with actual practice where the most precise control on geometry will generally be
achieved at the support region.

• For these spot check conditions, it was generally observed that bracing demands are more
critical for single-curvature bending conditions with less moment gradient (i.e., a smaller
Cb factor) than reverse-curvature conditions with significant moment gradient. This is
reflected in Eq. 2.3, in which the Cb factor of the unbraced segment of interest is in the
denominator.
Based on these results, it can be concluded that Eq. 2.3 is valid and generally conservative
for cases with reverse-curvature bending in noncomposite systems. That is to say, the bracing
strength requirements are not necessarily a function of positive or negative moment. For cases
with variable unbraced segment lengths and moment gradient factors, the bracing moment is to
be based on the brace point and unbraced segment that maximizes the Mr/CbLb component of
the equation regardless of which flange is in compression. In doing so, this ensures the critical
condition is covered. All cross-frame braces in the span would be subsequently designed for
this worst-case condition. These three variables are segment-specific, whereas the other vari-
ables and constants are consistent for the given span (e.g., the span length L is taken as the
full span length under consideration). One note to the above discussion is the often overly
conservative expression for Cb that is currently in AASHTO LRFD as of this writing. There is
significant work currently underway related to the moment gradient factor in the AASHTO
Specification that is likely to result in improved accuracy over the current Cb equation given
in the specification.
Note that Lb in Eq. 2.3 need not be taken as less than the maximum unbraced length permitted
for the girder segment based upon the required flexural strength, Mr. This criterion accounts for
scenarios in which a closer spacing of the braces is provided compared to the spacing required
to develop the required strength. In essence, this provision lessens the demands on cross-frames
for girders that would not buckle under the specified cross-frame layout (i.e., the girder would
partially or fully yield prior to buckling). In these instances, out-of-plane deformations would be
limited such that cross-frame bracing demands are relatively small. Hence, using the increased
Lb term in the denominator reflects that behavior.
As outlined in Section 3.5.1, bracing demands in the composite condition were deemed
less critical than in the noncomposite condition, given the lateral and torsional restraint pro-
vided by the concrete deck as well as the bridge bearings in the negative moment region. Thus,
the procedure introduced above is only necessary for the construction condition, where gird-
ers are far more susceptible to LTB. Note that, although cross-frame bracing requirements are
not critical for the composite condition, web stiffeners must be adequately proportioned to
prevent web distortion buckling effects in the negative moment regions per Yura (2001).
3.5.3 Bracing Stiffness Study

Similar to the bracing strength study previously outlined, a finite element parametric study
was developed and conducted to assess the applicability of the AISC bracing stiffness require-
ment (Eq. 2.1) for implementation into AASHTO LRFD. Second-order analyses were performed
on a variety of noncomposite bridge systems and bracing schemes to evaluate the torsional brace
stiffness required to limit out-of-plane girder deformations at critical buckling loads. Thus,
evaluating the relationship between internal girder moment and girder twist as a function
of the brace stiffness is the primary focus of this section. For clarity, only select results are
presented to highlight key trends that support the final suggestions and conclusions outlined
in Chapter 4. For a full overview of the results, refer to Appendix F.
Before presenting general results of the full data, a sample set is shown to illustrate key aspects
of the analytical results. Figure 3-40 presents the twist of an imperfect girder cross-section at
midspan (i.e., the critical location) as a function of the applied moment magnitude (also taken

at midspan) for the following select parameters: a single-span, twin-girder system (n g = 2)

comprised of Cross-section 2, one intermediate brace (n = 1), and uniformly distributed load-
ing on the top flange. As the applied load was increased, the out-of-plane girder deformations
resulting from second-order effects on the initial imperfection were recorded.
Along the x-axis, the cross-sectional twist of the girder at midspan, q, is normalized by the
initial imperfection, qo, as defined in Eq. 2.3. The q /qo ratio equals 1.0 at no applied load given
the assumed initial imperfection. Along the y-axis, the applied moment, M, is normalized by
the critical buckling capacity between brace points, Mcr. The applied moment for a single girder
is taken as the average internal moment in the multi-girder system. In some cases, the applied
moment does not reach Mcr due to in-plane stiffness effects. Instead, it is limited to 97% to 99%
Mcr. This response is plotted for twin-girder systems with various levels of cross-frame stiffness:
2, 2.5, 3, and 4 times the ideal stiffness, b i. Recall that the values for b i and Mcr are obtained from
eigenvalue analyses.
Many of these same plots were produced for different sets of parameters. The primary
goal of the plots is to determine the cross-frame stiffness that limits the q /qo ratio to 2.0 at
the critical buckling load (i.e., M/Mcr = 1.0). A ratio of 2.0 represents the case in which the
induced deformations at the buckling load are equal to the initial imperfection, which was the
basis for the AISC design procedures and Eq. 2.3.
From Figure 3-40, it is apparent that providing twice the ideal stiffness results in q /qo = 2.7 for
this particular example. The desired q /qo ratio of 2.0 is, however, achieved when providing three
times the ideal stiffness with even tighter tolerances achieved for a stiffer brace (i.e., four times
ideal stiffness). Additionally, a stiffer brace not only reduces the relative deformations in the
girders but also reduces the force demands in the brace members themselves, which impacts
the brace strength requirements discussed in Section 3.5.1. Similar observations were also made
for different sets of bracing and girder parameters.
Additional figures that investigate the effects of girder spacing, number of girders, load-
height effects, and intermediate bracing schemes are presented and summarized in Appendix F.
1.2
1.0
0.8
M / Mcr
0.6
βb = β2i
0.4 βb = 2.5 βi
βb = 3 βi
0.2 βb = 4 βi
0.0
1.0 1.5 2.0 2.5 3.0
θ / θo
Figure 3-40. Girder cross-sectional

twist as a function of applied uniform
moment for various brace stiffness
values (n = 1, Cross-section 2, and top
flange loading).

For clarity, these results are not included herein. However, the following items provide a
cursory overview of the major observations:
• The relative girder twist induced at critical buckling loads decreases as the brace stiffness
increases. In all cases, assigning stiffer cross-frame elements mitigates girder deformations.
• In general, adding intermediate braces (regardless of the stiffness) decreases the relative
cross-section twists.
• Increasing the girder spacing, particularly in twin-girder systems, not only improves the
system buckling capacity but also decreases the stiffness demands for the cross-frame braces.
Similarly, systems with increased girder redundancy generally resulted in an increase in the
in-plane stiffness of the bridge system, which effectively led to smaller girder deformations
at critical buckling loads.
• Top flange loading and the mid-height loading resulted in similar deformation behavior.
The effect of the transverse loading locations on the stiffness requirement for torsional
bracing was marginal.
• In general, assigning three times the ideal stiffness tends to result in girder cross-sectional
twists closer to two times qo than assigning twice the ideal stiffness.
Table 3-13 summarizes the torsional bracing stiffness requirements based on the results of
this parametric study. The final girder twist (as a ratio of the initial imperfection) is tabulated
for various cross-frame stiffness values, loading conditions, and girder cross-sections. Note
that these results correspond to twin girders only, because this system has been shown to
be most critical. For each q/qo ratio reported, the corresponding M/Mcr value is also shown. In
most cases, the applied moment reached the critical buckling moment (i.e., M/Mcr = 100%).
However, there were a few instances in which the critical buckling moment was not achieved;
in those cases, the girder twist at the maximum girder moment recorded is presented instead.
Table 3-13. Summary of stiffness requirements of torsional bracing for different

cross-sections and load conditions.
Cross- na 1 (Cb = 1.30) 3 (Cb = 1.06) 5 (Cb = 1.03)

Loading
frame
Type bf / d 1/6 1/4 1/3 1/6 1/4 1/3 1/6 1/4 1/3
Stiff., βb
θ / θo 3.15 2.72 2.89 3.59 2.74 3.15 2.7 2.55 3.29

2βi
M / Mcr 100% 100% 100% 98% 99% 99% 100% 100% 98%
θ / θo 2.77 2.32 2.46 2.9 2.27 2.64 1.96 2.03 3.2
Top 2.5βi
M / Mcr 100% 100% 100% 99% 100% 100% 100% 100% 99%
Flange
Loading θ / θo 2.56 2.11 2.22 2.39 1.94 2.21 1.71 1.86 3.14
3βi
M / Mcr 100% 100% 100% 100% 100% 100% 100% 100% 99%
θ / θo 2.3 1.89 1.97 1.8 1.67 1.91 1.49 1.68 3.07
4βi
M / Mcr 100% 100% 100% 100% 100% 100% 100% 100% 99%
θ / θo 3.41 2.92 3.11 3.29 2.61 2.85 2.42 2.4 3.51
2βi
M / Mcr 100% 100% 100% 99% 100% 100% 100% 94% 99%
θ / θo 2.98 2.47 2.63 2.41 2.03 2.35 1.9 2.05 3.4
Mid- 2.5βi
M / Mcr 100% 100% 100% 100% 100% 100% 100% 100% 99%
Height
Loading θ / θo 2.76 2.24 2.38 1.89 1.83 2.04 1.68 1.87 2.89
3βi
M / Mcr 100% 100% 100% 100% 100% 100% 100% 100% 100%
θ / θo 2.46 2 2.13 1.59 1.59 1.84 1.49 1.6 2.61
4βi
M / Mcr 100% 100% 100% 100% 100% 100% 100% 100% 100%
Notes:
a
Given the uniformly distributed loading (wL2/8 type loading), the moment gradient factor, Cb , is a function of the
bracing scheme.

In general, the table shows that providing twice the ideal stiffness (2b i) typically results in
load-induced girder twists exceeding the assumed twist in Eq. 2.3 (i.e., q/qo > 2). This implies
that the current assumption built into AISC Specifications (i.e., Eqs. 2.1 and 2.3) is invalid.
However, the cross-sectional twist can be reduced to magnitudes more consistent with the
assumed values by requiring braces to possess three times the ideal stiffness (3b i). Therefore,
for implementation into AASHTO LRFD, it is advised that Eq. 2.1 be modified to inherently
consider three times the ideal stiffness as a means to limit girder deformations. This is presented
mathematically by the following expression:
3.6 LM r2
b T,req = 3.3
f nEIyeff Cb2
A cross-frame must then be designed and detailed such that its total stiffness, as computed
by Eq. 2.2, exceeds the required stiffness from Eq. 3.3. Note that the constant in the equation
has increased 50% (i.e., from 2.4 to 3.6) to account for these recommended changes. This
modification not only impacts the brace stiffness requirements, but also the brace strength
requirements. Recall from Section 2.6 that the brace strength equations are effectively the
required brace stiffness multiplied by the assumed initial imperfection. With that in mind, it is
also recommended to modify Eq. 2.3 to account for the 3b i assumption outlined above. This
is reflected with the revised expression below:
0.036 M r L
M br = b Tq o = 3.4
nCb Lb
To determine the axial-force demands in the cross-frame members, the bracing moment
computed by the expression above can be resolved as a force couple to the top and bottom
nodes. This procedure is demonstrated in Appendices B and C. It is important to note that, as
an alternative, designers are also permitted to perform a large-displacement analysis to estimate
stability bracing force demands, provided that initial imperfections are considered (Helwig and
Yura 2015). For curved systems though, the effect of initial imperfections has been shown to
be less impactful given the curved geometry of the girders.
3.5.4 Consideration of Stability Forces in Design

As introduced in Section 1.1.1, cross-frames serve many functions throughout the construc-
tion and service life of a steel I-girder bridge. AASHTO LRFD Article 6.7.4.1 summarizes these
functions in the context of minimum design and analysis requirements. Despite not currently
providing any guidance on the topic, AASHTO LRFD includes stability bracing as one of the
design considerations for cross-frames. While Sections 3.5.1 and 3.5.3 examined different
strength and stiffness requirements to address this gap in design guidance, it is also important to
understand how these stability design forces interact with other cross-frame functions and load
cases. As such, this section provides an overview on how stability bracing strength requirements
can be considered in conjunction with wind loads, construction loads, etc.
As noted in the preceding sections, stability bracing requirements are most critical during
construction when the system is noncomposite. The critical construction stage usually occurs
during placement of the concrete deck since this results in the maximum moment that the
noncomposite steel girder must resist. The concrete deck, once hardened and composite with
the bridge girders, provides continuous restraint to the top flange and substantial bracing
benefits to the bottom flange in negative moment regions. As such, the discussion herein
is limited to force effects generated in cross-frame members during steel erection and deck

construction. Given that cross-frames are generally not required as stability braces in the
finished bridge, it is only necessary to evaluate stability-related force effects with other con-
struction force effects, including dead loads (e.g., wet concrete, formwork, fit-up forces),
construction loads (e.g., overhang construction, screed), and wind loading acting on the
bridge fascia. Cross-frame force effects related to the finished structure, such as live loads
and wearing surface dead loads, are then to be considered independently in their own set of
load combinations as currently established in AASHTO LRFD.
Because the stability-related force effects are critical during construction, at which point the
bridge presumably exhibits fully elastic behavior, the principle of superposition is applicable for
these specific load combinations. That is, the bracing strength forces computed from Eq. 3.4 or
a large-displacement analysis can be directly added to the force effects from dead, construction,
and wind loads derived from an elastic analysis.
Before introducing each load case, it is important to note that construction sequencing must
be considered when applying principles of superposition. Depending on the responsible party
at the specific construction condition, either the engineer-of-record (EOR) or the erector or
contractor’s engineer may need to evaluate stability bracing requirements at different stages of
construction. The behavior is often dependent on the state of cross-frame installation and/or
external bracing systems (i.e., guy cables, shoring, or holding cranes) at the specific erection
or construction stage. Therefore, the additive nature of these force effects is contingent on
consistent and concurrent bridge conditions. For example, stability bracing force effects related
to an intermediate phase of girder erection should not be combined with dead load force effects
related to wet concrete, as these two activities occur at different stages of construction. It is also
imperative that the EOR or the erector or contractor’s engineer maintain the proper sign when
summing the force effects of each load case (i.e., tension and compression).
With that in mind, dead loads consider both (i) internal force effects due to gravity loads and
(ii) “fit-up” forces due to externally applied loads by the erector to assemble the structural steel
during erection. The fit-up forces are largely dependent on which fit condition is selected by the
engineer and constructed by the steel erector (i.e., no-load fit, steel dead load fit, or total dead
load fit). Dead load force effects in cross-frames due to gravity loads on noncomposite systems
are commonly estimated using 2D or 3D analysis models. Many commercial design software
programs have also been designed and programmed to perform staged construction analysis
to handle the various states of stress throughout the construction process. This topic is covered
extensively in White et al. (2012). Guidance for estimating fit-up forces (when necessary),
on the other hand, is provided in AASHTO LRFD Article 6.7.2, which is largely based on the
work of White et al. (2015).
Construction-related force effects in cross-frames, such as those induced by overhang deck
construction, are often estimated with hand solutions as documented in Appendices B and C.
Lastly, wind loads on a steel superstructure during construction can be estimated with refined
analyses or the simplified procedures provided in AASHTO LRFD Article 4.6.2.7.
With the individual load cases outlined, the final step is related to load factors. Because
stability bracing force effects are a function of the factored girder moment (i.e., Mr in Eq. 3.4),
no additional load factors are needed when combining these effects with the other construction-
related force effects. If large-displacement analysis is used to estimate the bracing force demands,
then the designer must give special consideration to how load factors are applied in analysis and
design checks. The remainder of the load cases are subsequently factored based on the guidance
and tables provided in AASHTO LRFD Articles 3.4.1 and 3.4.2.
This entire process, as summarized above, is illustrated through two design examples in
Appendices B and C.


Based on the results of the Stability Study presented in the preceding subsections, there are
several conclusions that can be drawn with respect to the bracing requirements for cross-frames
and the LTB behavior of nonprismatic girders:
• Given that a composite deck provides continuous restraint to the top flange and substantial
restraint to the bottom flange, these bracing provisions are only necessary to evaluate during
the construction stages.
• In terms of implementation into AASHTO LRFD, it was determined that the general form
of the torsional brace strength equations from the 14th Edition of AISC (2010) are more
appropriate than the current 15th Edition (2016). The 15th Edition, which requires a design
brace moment of 2% of the maximum girder moment, was shown to significantly under-
predict the required brace moment. In contrast, the 14th Edition version of the strength
requirements is a function of the bracing layout and moment gradient factor. The proposed
expression, which is a modified version of the 14th Edition AISC equation, is provided as
Eq. 3.4 above.
• For cases with reverse-curvature bending, it is recommended to conservatively base the required
bracing moment on the unbraced segment that maximizes Mr /CbLb in Eq. 3.4. This ensures
that for each span (with a corresponding length, L, and number of intermediate braces, n),
the critical bracing moment is considered.
• Historically, design specifications have required engineers to provide torsional brace stiffness
equivalent to twice the ideal stiffness. This rule of thumb was developed by Winter (1960) and
was largely validated for columns rather than beams. Through FEA parametric studies, it was
concluded that providing three times the ideal stiffness better limits girder deformations and
cross-frame forces at critical buckling loads in beams. A proposed modification to the current
AISC approach is therefore provided in Eq. 3.3.
• Bracing strength demands, as computed by Eq. 3.4 or large-displacement analyses, can
be combined with other construction-related force effects via linear superposition. This
procedure is demonstrated through two design examples in Appendices B and C.
and commentary.

CHAPTER 4
Conclusions and Suggested Research
Up until the 1990s, specifications included a spacing limit that effectively standardized the
design of cross-frames. AASHTO LRFD Specifications have since eliminated a spacing limit
for straight bridges and instead require the design to be based on a rational analysis. Although
many bridge owners still utilize typical details for cross-frames, software tools have made it
possible for designers to develop improved predictions of cross-frame behavior and design
forces. However, to adequately make use of these software tools, there has become an increas-
ingly important need to modernize design guidance for cross-frames.
Although considerable research over the past several decades has improved cross-frame
design in the areas of stability bracing, load-induced fatigue resistance, erection fit-up, and
analysis for the noncomposite construction condition, the design industry has generally
lacked quantitatively based guidance on several other topics related to fatigue loading criteria,
analysis techniques, and stability bracing. In response, NCHRP Project 12-113 identified
several gaps in knowledge, including those generalized concepts as follows: (i) fatigue loading,
(ii) analysis, and (iii) stability bracing requirements of cross-frame systems in steel I-girder
bridges. These gaps in knowledge served as the primary objectives of this project, previously
identified as Objectives (a) through (e) in Chapter 1. To further clarify the intent of the
research, those five objectives were subsequently posed as a series of questions that have been
systematically addressed throughout the report.
This chapter summarizes the major findings from the experimental and analytical studies
performed in relation to those questions. Section 4.1 outlines the key conclusions of NCHRP
Project 12-113. Section 4.2 serves as the link between the major findings of the studies and the
proposed specification and commentary language provided in Appendix A. In other words,
the conclusions outlined in Section 4.1 are synthesized in the context of AASHTO LRFD, and
Section 4.2 explains the justification and rationale for the proposed language presented in the
appendix. Lastly, although this project significantly advances cross-frame analysis and design
practices, there are a number of related studies that merit additional investigation. Section
4.3 identifies research topics beyond the scope of NCHRP Project 12-113 that could further
improve the understanding of cross-frame behavior in steel I-girder systems.
4.1 Major Conclusions

As noted previously, this section summarizes the primary conclusions reached from the
experimental and analytical studies presented in Chapter 3. Given that the major research
questions introduced in Section 1.2 served as the organization throughout the report, the
conclusions provided herein are also organized in the same manner. Each research question,
including its overarching topic of interest (i.e., fatigue loading, analysis, or stability) and the
pertinent AASHTO LRFD articles, is presented in italicized font. Bulleted items, which were
134

Conclusions and Suggested Research 135
synthesized from the “Major Outcomes” sections in Chapter 3, are then provided to address
each question individually. Proposed changes to the corresponding AASHTO LRFD Articles are
subsequently presented in Section 4.2 based on the commentary herein.
[Fatigue loading – Articles 3.6.1.4, 6.6.1.2.1] Is the current fatigue load model in terms of truck
position (i.e., single design truck passages positioned in various transverse lane positions) appro-
priate for cross-frame analysis and design? Or do multiple presence effects need to be considered?
• The WIM study (Section 3.2) confirmed that the “dual truck” event, which maximizes force
reversal in critical cross-frames and was initially considered in the 7th Edition AASHTO
Specifications, is a rare occurrence. As such, the current 9th Edition load criteria (i.e., a single
design truck positioned in all longitudinal and transverse positions with the truck confined
to one critical transverse position per each longitudinal position) is more appropriate.
• In evaluating the AASHTO fatigue criteria, it was concluded that additional economies are
possible for cross-frame design. This is largely attributed to (i) the sensitivity of cross-frame
response to transverse load position and (ii) the current fatigue load factors. The fatigue load
factors introduced in the 8th Edition AASHTO LRFD, which were calibrated based on girder
response, do not accurately represent the response of cross-frames. As such, one or both of
these items above could be addressed to maximize the efficiency of the design criteria.
• In terms of item (i) above, current design criteria require all possible lane positions be
considered, regardless of design lanes or actual lane striping. A more realistic scenario would
be to apply fatigue loading only in anticipated drive lanes. Item (ii) and cross-frame-specific
load factors are addressed with the next question.
[Fatigue loading – Article 3.4.1] Are the current AASHTO LRFD Fatigue I and II load factors,
which were calibrated for girder force effects and recent WIM data, appropriate for cross-frame
analysis and design?
• The WIM study summarized in this report indicates that the existing load factors of 1.75 for
Fatigue I and 0.8 for Fatigue II are conservative for the fatigue limit state design of cross-
frames by at least 35%, specifically when the AASHTO fatigue design truck is placed in every
possible transverse position during analysis and design. To improve economy in the AASHTO
LRFD guidance, this study demonstrated that either the fatigue design truck could be posi-
tioned in striped lanes (rather than permitting design trucks in unrealistic lane positions
such as adjacent to the bridge barriers), or cross-frames could be designed to adjusted load
factors. Based on feedback from the NCHRP Project 12-113 panel, it was deemed more
desirable to propose adjusted cross-frame-specific load factors.
• As an alternative to introducing two new load factors for the Fatigue I and II limit states for
cross-frames, it is possible to apply a single adjustment factor to the existing load factors,
similar to the current use of an adjustment factor in the fatigue design of orthotropic decks in
AASHTO LRFD Article 3.4.4.
• Using a single adjustment factor of 0.65 applied to the Fatigue I load factor of 1.75 (i.e.,
a resultant load factor for cross-frames of 1.14) and the Fatigue II load factor of 0.8 (i.e.,
a resultant load factor of 0.52), this study demonstrated that the resulting reliability indices
calculated via Monte Carlo simulation satisfy a minimum assumed target reliability of unity,
as recommended in SHRP 2 R19B (Modjeski and Masters 2015).
[Fatigue loading – Table 6.6.1.2.5-2] Is the “n” value (i.e., number of cycles per truck passage)
currently assumed for the generic “transverse member” designation applicable for cross-frames?
• The reduced fatigue load factors specific to cross-frames proposed in the preceding question
are calibrated assuming the number of cycles per design truck passage, n, is equal to unity,
regardless of cross-frame spacing. This is consistent with the analytical and experimental
analyses conducted in this study.

[Fatigue loading – Article 6.7.4.2] What is the influence of composite bridge geometry (e.g., support
skew, horizontal curvature) and cross-frame layout (e.g., cross-frame spacing, staggered layout)
on the load-induced fatigue behavior of cross-frame systems (including governing force effects,
critical lane loading, and critical members under AASHTO fatigue loading criteria)?
• In general, cross-frame response is sensitive to transverse truck placement. The sensitivity
of cross-frame response to longitudinal truck placement depends heavily on support skew
and horizontal curvature. Similarly, critical lane position also depends on a variety of
parameters. For instance, truck passages closer to the outer edges of the deck tend to
maximize cross-frame forces in skewed and curved bridges, but not in straight bridges with
normal supports.
• In terms of implementation into AASHTO LRFD, the results of this study indicate that the
current specification language in Article 3.6.1.4.3a, “a single design truck shall be positioned
transversely and longitudinally to maximize stress range at the detail under consideration,”
is the best way to ensure the critical load position is considered by the designer. This require-
ment is best executed with an influence-surface analysis. Thus, performing an influence-
surface analysis for the entire bridge deck, regardless of design or striped lanes, in conjunction
with the proposed cross-frame-specific fatigue load factors more accurately represents the
response of cross-frames to live load traffic.
• The governing cross-frame member in a given bridge depends on many variables, including
girder spacing, support skew, and striped lane positions. Bottom struts tend to see the
most substantial load-induced force effects in composite systems, especially those near
the supports of heavily skewed bridges. However, to ensure the critical force effects are
considered in design, an influence-surface analysis is necessary (i.e., to examine all possible
truck positions and cross-frame force effects).
• In general, heavily skewed and/or heavily curved bridges, particularly those with contiguous
lines of braces, generally produce the most significant force effects in cross-frame members.
To avoid an iterative “chase-your-tail” design solution, whereby the designer uses increasingly
large trial sizes for cross-frame members which in turn attract larger cross-frame forces, other
key bridge parameters can be adjusted to mitigate force effects. For instance, using a dis
continuous, staggered cross-frame layout is a practical and economical solution for skewed
systems. Increasing the deck thickness or decreasing girder spacing can also reduce cross-
frame forces. However, these measures have significant impacts on the rest of the design and
may therefore be less economical.
[Fatigue loading – Articles 6.7.4.1, 6.7.4.2] Is it necessary to perform a refined analysis, either
simplified 2D or 3D methods, for straight and non-skewed bridges? In other words, are the mini-
mum design requirements outlined in Article 6.7.4.1 appropriate?
• When comparing the response of cross-frames to measured WIM records, it was observed
that load-induced fatigue forces are most significant in skewed and/or curved bridges, for
which Ic > 1, Is > 0.15, or Ic > 0.5 and Is > 0.1. For bridges that do not fall into one of these
categories, a refined 2D or 3D analysis is likely not warranted. In these instances, other
design considerations such as stability bracing during construction often govern the design.
Therefore, the minimum design requirements with respect to straight bridges with normal
supports in Article 6.7.4.1 (in addition to stability bracing requirements) are appropriate.
[Analysis – Article 4.6.3.3.4] Is the current established R-factor (0.65AE), which was based on
analytical and experimental studies of a noncomposite system, appropriate for cross-frames in the
composite condition? Are there alternative 3D modeling approaches for cross-frames?
• 3D commercial design software programs commonly represent cross-frames as pin-ended
truss elements, which inherently misrepresent the in-plane and out-of-plane stiffness of

the system. The stiffness modification approach accounts for these limitations by adjusting
the axial stiffness of the cross-frame members in the analysis model. The proposed eccentric-
beam approach considers flexural deformations by explicitly representing the eccentric load
path through the cross-frame connections.
• Cross-frame deformational response in composite systems (i.e., in-service) is different than
the response in noncomposite systems (i.e., during construction). As such, the established
R-factor in AASHTO LRFD that was developed specifically for the noncomposite condition
should be reconsidered for composite conditions and live load force effects.
• Independent stiffness modification factors are proposed for the construction stages (Rcon = 0.65)
and for in-service conditions (Rser = 0.75). Note that the appropriate R-factor to be assigned
for the in-service condition is largely a function of bridge geometry, cross-frame details,
and uncertain loading conditions. Although scatter was observed in the results, the proposed
R-factor is a simple solution to a complex problem that produces reasonably accurate
approximations of the true cross-frame stiffness. Considering that many designers often
prefer simple alternatives over sophisticated refined analyses, it serves an important role
in AASHTO LRFD guidance moving forward.
• Analytical results demonstrated that the proposed eccentric-beam analysis method improves
the repeatability and reliability of 3D models by eliminating several sources of uncertainty
associated with the R-factor approach. Although the R-factor approach serves a vital purpose
in practice due to its ease of use, the proposed eccentric-beam method offers an alternative
to engineers seeking a more refined solution.
[Analysis – Article 4.6.3.3.2] What are the limitations of simplified 2D analysis techniques
commonly used by popular commercial bridge design software programs in terms of predicting
cross-frame force effects in composite systems? Are there methods available to improve these
simplified techniques?
• In general terms, designers must be cognizant of the potential trade-offs between sophis-
ticated and simplified analyses. A central theme explored throughout this report is related
to the balance between increased computational complexity and improved reliability versus
simplified modeling and improved ease of use. The results presented in this report clearly
highlight this trade-off. The conclusions below are intended to provide practicing engineers
the technical and quantitative background to help them make informed decisions, while not
necessarily advocating one method or the other for a specific project need.
• Although 3D models are more time-consuming to develop than 2D counterparts, these
models offer solutions with improved accuracy, reliability, and repeatability. 2D analysis
methods have traditionally been shown to produce reasonable approximations of girder
force effects, but this study demonstrated the significant limitations of 2D analyses asso
ciated with cross-frame force effects due to truck loads.
• The traditional approach of connecting cross-frame members into a shared node along
the web-to-flange juncture in 3D models is also acceptable, as long as the true stiffness is
approximated with R-factors or the eccentric-beam approach.
• 2D PEB models generally produce more accurate cross-frame results than 2D grillage
models because the concrete deck is explicitly considered. Although there are methods to
consider the effective transverse stiffness of the deck, it is more accurate to represent the
deck as a thin shell element that is rigidly connected to the girder to simulate composite
action.
• Common postprocessing practices for 2D methods tend to produce erroneous results for
top strut members and potentially inaccurate results for diagonal members in X-frames,
regardless of bridge type. If a designer elects to use a 2D modeling approach for these
complex cases, it is recommended to incorporate simple modifications to the model
(e.g., utilizing the Timoshenko approach for equivalent cross-frame beams) and to the

postprocessing (e.g., considering various shear force distributions in the design of X-frame
diagonals or considering the contributions of the concrete deck when evaluating grillage
model results similar to Figure 2-17). Still, even after implementing these improvements to
2D analyses, substantial error is likely in cross-frame force predictions, especially in bridges
with significant horizontal curvature and/or support skews.
• Based on the conclusions above, it is recommended that cross-frame force effects—
particularly those related to live loads—for any bridge satisfying the skew and connectivity
index limits established above should be evaluated using 3D modeling techniques. In general
terms, this implies that obtaining live load cross-frame forces for heavily skewed and/or
curved systems is best suited for 3D analysis. Although the focus of this study is related to the
fatigue limit state, this statement is also directly applicable to the strength limit state design
(and associated analysis) of cross-frames.
[Stability bracing – Article N/A] Can the AISC design guidelines for stability bracing be incorporated
into AASHTO LRFD? Are special requirements needed for negative moment regions of continuous
systems?
• Given that a composite deck provides continuous restraint to the top flange and substantial
restraint to the bottom flange, cross-frame bracing requirements are only necessary to be
evaluated for construction conditions.
• Through FEA parametric studies, it was concluded that providing three times the ideal
stiffness for cross-frames better limits girder deformations and cross-frame forces at critical
buckling loads. As such, a proposed modification to the current AISC expression was
provided in Eq. 3.3 for implementation into AASHTO LRFD.
• The general form of the torsional bracing strength equations from the 14th Edition of AISC
(2010) is more appropriate than the current 15th Edition (2016) for steel bridge applications.
Based on the revisions to the torsional bracing stiffness requirements above, Eq. 3.4 was
proposed for implementation into AASHTO LRFD.
• For cases with reverse-curvature bending, it is recommended to conservatively base the
required bracing moment on the unbraced segment that maximizes Mr /CbLb in Eq. 3.4. This
ensures that, for each span (with a corresponding length, L, and number of intermediate
braces, n), the critical bracing moment is considered.
[Stability bracing – Article N/A] How are these stability bracing requirements combined with other
load conditions such as wind?
• Bracing strength demands, as computed by Eq. 3.4 or from a large-displacement analysis,
can be combined with other construction-related force effects via linear superposition. These
bracing forces must be evaluated at all stages of steel erection and deck construction. It
is also important that the designer evaluates these load cases under the same concurrent
construction conditions. For example, stability bracing force effects related to an intermediate
phase of girder erection should not be combined with dead load force effects related to wet
concrete, as these two activities occur at different stages of construction. Because stability
bracing force effects are a function of the factored girder moment (i.e., Mr in Eq. 3.4), no
additional load factors are needed when combining these effects with the other construction-
related force effects.
4.2 Suggestions for Implementation

Based on the major conclusions outlined above, this section highlights the articles in the
AASHTO LRFD Specifications to be modified. In general, the rationale and justification for
the proposed specification and commentary language presented in Appendix A is explained.

In terms of organization, each specific AASHTO LRFD Article is listed in italicized font herein
followed by bulleted items that expound on the proposed modifications.
Articles 3.4.1 and 3.4.5 (proposed)
• As noted in Section 4.1, cross-frame-specific load factors that more accurately represent the
live load response of cross-frames to the U.S. truck spectra have been proposed. Rather
than introduce two new Fatigue I and II load factors, an additional adjustment factor (0.65)
to be applied to the existing load factors is proposed, similar to the approach taken for the
fatigue design of orthotropic decks in AASHTO LRFD Section 3.4.4. As such, a new article is
proposed (Article 3.4.5) to introduce this additional adjustment factor.
Article 3.6.1.4
• As noted in Section 4.1, conservatism in the current AASHTO fatigue loading model has been
eliminated by proposing reduced cross-frame-specific fatigue load factors as opposed to
limiting design trucks to a specified lane. As such, the findings of this study concur with and
validate the current commentary language in this article, which addresses the uncertainty
of future lane striping changes and deck widening. Thus, no additional specification or
commentary language is recommended.
Article 4.6.1.2.4
• Article 4.6.1.2.4 currently specifies that cross-frames and diaphragms in straight and slightly
curved I-girder bridges shall be designed, at a minimum, for wind loads and slenderness
requirements. Based on the findings of the Stability Study, it is recommended to include
stability bracing as an additional minimum design requirement for these I-girder bridges,
especially during deck construction. As such, a reference to proposed Article 6.7.4.2.2 is
added to the specification language in this article.
Article 4.6.3.3.2
• Based on the findings of this study, the commentary language in Article C4.6.3.3.2 with regards
to equivalent torsional stiffness to account for warping should be maintained. It was observed
that neglecting warping-torsional rigidity in composite bridges is much less impactful than
in noncomposite bridges as documented by White et al. (2012). However, many designers
utilize commercial design software packages that perform staged construction. Thus, it is
beneficial to incorporate warping rigidity through all stages of construction and service, even
if it does not significantly influence the cross-frame response to live loads.
• As noted in Section 4.1, 2D PEB and grillage model approaches may lead to more significant
errors in the prediction of cross-frame forces regardless of bridge geometry. This was largely
attributed to the postprocessing procedures commonly implemented that convert equiva-
lent beam shear and moments into cross-frame member forces. Thus, commentary language
has been added to help designers make informed decisions regarding analysis for their
project needs.
Article 4.6.3.3.4
• Article 4.6.3.3.4 has been divided into three separate sub-articles to clarify the differences
between 2D and 3D modeling techniques of cross-frames. Proposed Article 4.6.3.3.4a
focuses on current 9th Edition specification and commentary language as well as proposed
language with respect to 2D analysis. Proposed Article 4.6.3.3.4b outlines 3D analysis
methods of cross-frames, including the traditional truss-element modeling approach and the
proposed eccentric-beam approach in the commentary. Lastly, proposed Article 4.6.3.3.4c
discusses the use of stiffness modification factors to be considered in both 2D and 3D models
of cross-frames.

Article 4.6.3.3.4a (proposed)

• For 2D modeling of cross-frames, the proposed postprocessing tools that improve the
accuracy of the force predictions are introduced with figures, as are analysis strategies that
consider the transverse stiffness effects of the concrete deck in grillage-type models.
Article 4.6.3.3.4b (proposed)
• The eccentric-beam modeling approach is introduced in the commentary as an alternative
to the truss-element (with R-factors) approach. As presented in this report, the eccentric-
beam approach encompasses a wide range of modeling assumptions with varying levels of
accuracy and modeling simplicity. Three specific variations were outlined and accompanied
by sample calculations in Appendix F, but the possible set of assumptions is certainly not
limited to those documented here. As a result, the reader can be referred to the final report
of NCHRP Project 12-113 for guidance, if desired, but detailed instructions on a specific
method are not offered in the specifications.
Article 4.6.3.3.4c (proposed)
• The stiffness response of cross-frames in noncomposite systems for stability-related deforma-
tion patterns should be clearly differentiated from the response of cross-frames in composite
systems subjected to highly variable live loads. This is accomplished by establishing a separate
stiffness modification factor for in-service conditions (Rser = 0.75) compared to the previous
recommendation of Rcon = 0.65 that was developed for noncomposite steel girders during
construction. These factors are applied to both 2D and 3D cross-frame analysis models.
• Currently, Article 4.6.3.3.4 references end connection eccentricities as the primary moti-
vation for the stiffness modification factors. However, as discussed in this report, these
modification factors inherently account for other effects beyond just eccentric end connec-
tions, including in-plane rotational restraint provided by the connection and gusset plates.
Accordingly, the proposed specification updates broaden the language by removing the
references to “end connection eccentricities” such that the reader understands the full
implications of using the simplified truss-element modeling approach.
Articles 6.6.1.2.1 and 6.6.1.2.2
• Based on the WIM study, it was confirmed that the removal of the “double truck case”
in the 2016 Interims to the 7th Edition of the specifications was warranted. Given that
Article C6.6.1.2.1 addresses this properly, no substantial changes to the specification or
commentary language are recommended. However, it is suggested to move this commentary
language to Article C6.6.1.2.2, along with providing additional guidance on how factored
cross-frame fatigue stresses are to be computed based on computer analysis results.
Table 6.6.1.2.5-2
• As noted in Section 4.1, the proposed adjustments to the Fatigue I and II load factors for
cross-frame design were calibrated based on the assumption of n = 1 (i.e., one stress cycle
per truck passage) given the observations made from experimental and analytical studies.
As such, the ambiguous designation of “transverse members” is clarified by differentiating
between cross-frames and floorbeam members. The n-value to use for cross-frames is defined
as 1.0 and is independent of longitudinal spacing.
Article 6.7.4.1
• Given that (i) live load cross-frame forces in geometrically simple bridges are generally
small, and (ii) 3D analysis models, although more accurate, can be computationally inten-
sive, it is important to establish bounds for when refined analyses are required to obtain
cross-frame design forces from gravity loads (e.g., live loads). As such, the skew and

connectivity index limits established previously (Ic > 1, Is > 0.15, or Ic > 0.5 and Is > 0.1)
are introduced in the commentary language to expound on the minimum design consid-
erations in this article. These limits also complement the general approach and language
in Article 4.6.3.3.2.
• In general, this proposed language helps clarify the design considerations and what consti-
tutes a “rational analysis.”
Article 6.7.4.2
• Article 6.7.4.2 is divided into two independent subheadings based on the addition of
stability bracing requirements for cross-frames. The specification and commentary language
in the original Article 6.7.4.2 are now contained in proposed Article 6.7.4.2.1. Stability bracing
requirements are provided in proposed Article 6.7.4.2.2.
Article 6.7.4.2.1 (proposed)
• Similar to the discussion for Article 4.6.1.2.4, the minimum design requirements for cross-
frames in I-girder bridges should include the applicable stability bracing checks at all critical
stages of construction. Specification language is added accordingly.
• This article currently highlights good design and detailing practices of cross-frames, particu-
larly in skewed and/or curved systems. Given that contiguous versus discontinuous cross-
frame lines are already discussed in the commentary with respect to noncomposite systems,
the major findings related to the Fatigue Loading Study are best suited here.
• As documented in Section 4.1, cross-frame force effects are highly variable and depend on
a number of parameters. Ultimately, the two most important aspects of the findings are
that: (i) bottom struts and diagonal members near supports tend to govern fatigue design in
skewed and/or curved bridges and (ii) there are a number of ways that designers can mitigate
design forces in cross-frame elements other than increasing the cross-sectional area. The
most practical and economical way to mitigate load-induced cross-frame forces in skewed
systems is to arrange the braces in a discontinuous, staggered layout. With that in mind,
proposed commentary language is provided to emphasize these findings.
Article 6.7.4.2.2 (proposed)
• On the specification side, Eqs. 3.3 and 3.4 are included, which present the required stiff-
ness and strength of cross-frames or diaphragms serving as torsional braces, respectively.
Additionally, it is important to include guidance on how the total stiffness of a cross-frame
system is calculated (i.e., Eq. 2.2). As such, guidance from Yura (2001) and specifically the
equations to estimate the stiffness of the torsional brace considering connection flexibility,
in-plane girder effects, and cross-sectional distortion effects (also documented in the AISC
Specifications) are provided for reference. The geometric limits, at which in-plane girder
stiffness and cross-section distortion effects become negligible, are also established.
• On the commentary side, much of the information included in the AISC Specifications is
adapted given that it directly applies to steel bridge applications. For instance, the basis of the
original strength and stiffness requirements (Taylor and Ojalvo 1966) is outlined, as is the
critical shape imperfection (Wang and Helwig 2005). Furthermore, expressions that consider
the effect of skewed cross-frames relative to a line normal to the longitudinal axis of the bridge
girders is provided based on work conducted by Wang and Helwig (2008).
• Given the bracing strength requirements specified in proposed Article 6.7.4.2.2 herein, it is
important to provide designers with guidance on how to utilize these forces in the context
of strength limit state design for all stages of construction. As noted in Section 4.1, stability
bracing force effects are to be combined with other construction-related load cases via
principles of superposition. Additional specification and commentary language have been
proposed to explain this procedure as well as the pertinent load factors. Design examples in
Appendices B and C demonstrate these calculations.

4.3 Future Research Needs

Although this project significantly advances cross-frame analysis and design practices in a
wide range of areas, there are a number of related studies that merit additional investigation.
These additional research topics are identified below.
• Investigate the benefits of lean-on braces in terms of relieving load-induced force effects in
composite skewed bridge systems. Although lean-on braces were examined during the field
experimental phase of NCHRP Project 12-113, a more comprehensive review of lean-on
brace response to truck traffic is warranted, especially as more lean-on systems continue to be
implemented in practice across the United States.
• Examine the stiffness implications of eccentric end connections on cross-frame systems utilizing
tee sections. This study focused primarily on single angles, as they represent the most common
steel section used in cross-frames across the Unites States. However, additional research on
the behavior of tee-section cross-frame members in composite systems is needed.
• Investigate methods to improve the prediction of cross-frame forces in 2D analysis software.
As documented in this report, current 2D bridge design software can provide potentially
erroneous predictions of cross-frame forces in composite bridges. Considering the popularity
of 2D bridge design software, development of methods to provide more accurate cross-frame
forces would be highly beneficial to bridge designers.
• Experimentally investigate the effects of staggered cross-frame layouts on lateral flange stresses
in skewed and curved bridge systems. Although substantial analytical efforts have been made
with regards to lateral flange stresses in girders (White et al. 2012), the research team is
unaware of any field or laboratory studies to examine the long-term fatigue ramifications of
these stresses.
• Develop innovative details to improve the fatigue category and reduce the cost of cross-frames.
While this report focused primarily on the fatigue loading characteristics of cross-frames,
additional studies on the nominal fatigue resistance of welded cross-frame details could be
very beneficial to bridge designers.

References
Abaqus/CAE. 2017. Abaqus/CAE User’s Guide.

American Association of State Highway and Transportation Officials/National Steel Bridge Alliance. 2019.
G13.1 Guidelines for Steel Girder Bridge Analysis, 3rd ed.
———. 2017. LRFD Bridge Design Specifications, 8th ed. Washington, D.C.
———. 1994. LRFD Bridge Design Specifications, 1st ed. Washington, D.C.
———. 2020. LRFD Bridge Design Specifications, 9th ed. Washington, D.C.
American Institute of Steel Construction. 2010. Specification for Structural Steel Buildings, 14th ed. Chicago, IL.
———. 2016. Specification for Structural Steel Buildings,15th ed. Chicago, IL.
Barth, K. 2015. Steel Bridge Design Handbook Design Example 2A: Two-Span Continuous Straight Composite
Steel I-Girder Bridge. Washington, D.C.: Federal Highway Administration.
Battistini, A., W. Wang, S. Donahue, T. Helwig, M. Engelhardt, and K. Frank. 2013. Improved Cross Frame Details.
TxDOT 0-6564 Final Report, Texas Department of Transportation.
Battistini, A., W. Wang, T. Helwig, M. Engelhardt, and K. Frank. 2016. “Stiffness Behavior of Cross Frame in Steel
Bridge Systems.” Journal of Bridge Engineering 21 (6).
Chavel, B., D. Coletti, K. Frank, M. Grubb, B. McEleney, R. Medlock, and D. White. 2016. Skewed and Curved Steel
I-Girder Bridge Fit. Chicago, IL: National Steel Bridge Alliance.
Connor, R., and J. Fisher. 2006. “Identifying Effective and Ineffective Retrofits for Distortion Fatigue Cracking
in Steel Bridges Using Field Instrumentation.” Journal of Bridge Engineering (American Society of Civil
Engineers) 11 (6): 745–752.
Downing, S., and D. Socie. 1982. “Simple Rainflow Counting Algorithms.” International Journal of Fatigue
(Butterworth & Co) 31-40.
Egilmez, O., T. Helwig, and R. Herman. 2016. “Using Metal Deck Forms for Construction Bracing in Steel
Bridges.” Journal of Bridge Engineering (American Society of Civil Engineers) 21 (5): 1–12.
Fasl, J. 2013. Estimating the Remaining Fatigue Life of Steel Bridges Using Field Measurements. PhD diss., The
University of Texas at Austin.
Fisher, J. W., J. Jin, D. C. Wagner, and B. T. Yen. 1990. NCHRP Research Report 336: Distortion-Induced Fatigue
Cracking in Steel Bridges. Washington, DC: Transportation Research Board.
Han, L., and T. Helwig. 2016. “Effect of Girder Continuity and Imperfections on System Buckling of Narrow
I-Girder Systems.” Proceedings of the Structural Stability Research Council. Orlando, FL: Structural Stability
Research Council.
———. 2017. “Nonlinear Behavior of Global Lateral Buckling of I-Girder Systems.” Proceedings of the Annual
Stability Conference. San Antonio, TX: Structural Stability Research Council.
Hartman, A., H. Hassel, C. Adams, C. Bennett, A. Matamoros, and S. Rolfe. 2010. “Effects of Cross-Frame
Placement and Skew on Distortion-Induced Fatigue in Steel Bridges.” Transportation Research Record:
Journal of the Transportation Research Board 2200 (1): 62–68.
Hassel, H., C. Bennett, A. Matamoros, and S. Rolfe. 2013. “Parametric Analysis of Cross-Frame Layout on
Distortion-Induced Fatigue in Skewed Steel Bridges.” Journal of Bridge Engineering (American Society of
Civil Engineers) 18 (7): 601–611.
Helwig, T., and J. Yura. 2015. Bracing System Design. Steel Bridge Design Handbook, Publication No. FHWA-
HIF-16-002 - Vol. 13, U.S. Department of Transportation, Federal Highway Administration.
Helwig, T., and L. Wang. 2003. Cross-Frame and Diaphragm Behavior for Steel Bridges with Skewed Supports.
Research Report 1772-1, Austin, TX: Report for Texas Department of Transportation.
Helwig, T., and Z. Fan. 2000. Field and Computational Studies of Steel Trapezoidal Box Girder Bridges. TxDOT
Research Report 1395-3, Houston, Texas: The University of Houston.
143

Helwig, T., K. Frank, and J. Yura. 1997. “Lateral-Torsional Buckling of Singly Symmetric I-Beams.” Journal of
Structural Engineering 123 (9): 1172–1179.
Keating, P., D. Mertz, J. Kulicki, and C. Hess. 1990. Economical and Fatigue Resistant Steel Bridge Details.
FHWA-H1-90-043, Washington, D.C.: Federal Highway Administration.
Kulicki, J., Z. Prucz, C. Clancy, D. Mertz, and A. Nowak. 2007. Final Report, NCHRP Project 20-7/186,
“Updating the Calibration Report for AASHTO LRFD Code.” Washington, D.C.: Transportation Research
Board.
Liu, Y., and T. Helwig. 2020. “Torsional Brace Strength Requirements for Steel I-Girder Systems.” Journal of Bridge
Engineering (American Society of Civil Engineers) 146 (1): 1–11.
McDonald, G., and K. Frank. 2009. The Fatigue Performance of Angle Cross-Frame Members in Bridges. MS Thesis,
The University of Texas at Austin.
Modjeski and Masters. 2015. SHRP 2 Report S2-R19B-RW-1: Bridges for Service Life Beyond 100 Years: Service
Limit State Design. Washington, D.C.: Transportation Research Board of the National Academies.
Nowak, A. 1999. NCHRP Research Report 368: Calibration of LRFD Bridge Design Code. Washington, D.C.:
Transportation Research Board.
Prado, E., and D. White. 2015. “Assessment of Basic Steel I-Section Beam Bracing Requirements by Test
Simulation.” Report to the Metal Building Manufacturers Association.
Quadrato, C., A. Battistini, T. Helwig, M. Engelhardt, and K. Frank. 2014. “Increasing Girder Elastic Buckling
Strength using Split Pipe Bearing Stiffeners.” Journal of Bridge Engineering (American Society of Civil
Engineers) 19 (4).
Reichenbach, M., Y. Liu, T. Helwig, and M. Engelhardt. 2020. “Lateral-Torsional Buckling of Singly Symmetric
I-Girders with Stepped Flanges.” Journal of Structural Engineering (American Society of Civil Engineers).
Rivera, J., and B. Chavel. 2015. Steel Bridge Design Handbook Design Example 3: Three-Span Continuous Hori-
zontally Curved Composite Steel I-Girder Bridge. Washington, D.C.: Federal Highway Administration.
Romage, M. 2008. Field Measurements on Lean-On Bracing System for Steel I-Girder Bridges with Skewed Supports.
Thesis, The University of Texas at Austin.
Sivakumar, B., M. Ghosn, and F. Moses. 2011. NCHRP Research Report 683: Protocols for Collecting and Using
Traffic Data in Bridge Design. Washington, D.C.: Transportation Research Board.
Tadros, M., N. Al-Omaishi, S. Seguirant, and J. Gallt. 2003. NCHRP Research Report 496: Prestress Losses in
Pretensioned High-Strength Concrete Bridge Girders. Washington, D.C.: Transportation Research Board.
Taylor, A., and M. Ojalvo. 1966. “Torsional Restraint of Lateral Buckling.” Journal of the Structural Division
(American Society of Civil Engineers) 92 (ST2): 115–129.
Texas Department of Transportation. 2020. Transportation Planning Maps. https://www.txdot.gov/inside-txdot/
division/transportation-planning/maps.html.
Timoshenko, S., and Gere, J. 1961. Theory of Elastic Stability. New York: McGraw-Hill.
Wang, L., and T. Helwig. 2005. “Critical Imperfections for Beam Bracing Systems.” Journal of Structural Engi-
neering (American Society of Civil Engineers) 131 (6): 933–940.
Wang, L., and T. Helwig. 2008. “Stability Bracing Requirements for Steel Bridge Girders with Skewed Supports.”
Journal of Bridge Engineering (American Society of Civil Engineers) 13 (2): 149–157.
Wang, W. 2013. A Study of Stiffness of Steel Bridge Cross Frames. PhD diss., The University of Texas at Austin.
White, D., D. Coletti, B. Chavel, A. Sanchez, C. Ozgur, J. Jimenez Chong, R. Leon, et al. 2012. NCHRP Research
Report 725: Guidelines for Analysis Methods and Construction Engineering of Curved and Skewed Steel Girder
Bridges. Washington, D.C.: Transportation Research Board.
White, D., T. Nguyen, D. Coletti, B. Chavel, M. Grubb, and C. Boring. 2015. Final Report, NCHRP 20-07/
Task 355, “Guidelines for Reliable Fit-Up of Steel I-Girder Bridges.” Washington, D.C.: Transportation
Research Board.
White, J. 2020. Evaluation of Fatigue Design Load Models for Cross-Frames in Steel I-Girder Bridges. PhD diss.,
The University of Texas at Austin.
Winter, G. 1960. “Lateral Bracing of Columns and Beams.” Journal of Structural Engineering (American Society
of Civil Engineers) 125: 809–825.
Yura, J. 2001. “Fundamentals of Beam Bracing.” Engineering Journal (American Institute of Steel Construction)
11–26.
Yura, J., B. Philips, S. Raju, and S. Webb. 1992. Bracing of Steel Beams in Bridges. Research Report 1239-4F, Austin,
TX: Center for Transportation Research, University of Texas at Austin.
Yura, J., T. Helwig, R. Herman, and C. Zhou. 2008. “Global Lateral Buckling of I-Shaped Girder Systems.” Journal
of Structural Engineering (American Society of Civil Engineers) 134 (9): 1487–1494.

APPENDIX A
Proposed Modifications
to AASHTO LRFD
The proposed modifications to AASHTO LRFD will be published by AASHTO.
A-1


Abbreviations and acronyms used without definitions in TRB publications:

A4A Airlines for America
AAAE American Association of Airport Executives
AASHO American Association of State Highway Officials
AASHTO American Association of State Highway and Transportation Officials
ACI–NA Airports Council International–North America
ACRP Airport Cooperative Research Program
ADA Americans with Disabilities Act
APTA American Public Transportation Association
ASCE American Society of Civil Engineers
ASME American Society of Mechanical Engineers
ASTM American Society for Testing and Materials
ATA American Trucking Associations
CTAA Community Transportation Association of America
CTBSSP Commercial Truck and Bus Safety Synthesis Program
DHS Department of Homeland Security
DOE Department of Energy
EPA Environmental Protection Agency
FAA Federal Aviation Administration
FAST Fixing America’s Surface Transportation Act (2015)
FHWA Federal Highway Administration
FMCSA Federal Motor Carrier Safety Administration
FRA Federal Railroad Administration
FTA Federal Transit Administration
HMCRP Hazardous Materials Cooperative Research Program
IEEE Institute of Electrical and Electronics Engineers
ISTEA Intermodal Surface Transportation Efficiency Act of 1991
ITE Institute of Transportation Engineers
MAP-21 Moving Ahead for Progress in the 21st Century Act (2012)
NASA National Aeronautics and Space Administration
NASAO National Association of State Aviation Officials
NCFRP National Cooperative Freight Research Program
NCHRP National Cooperative Highway Research Program
NHTSA National Highway Traffic Safety Administration
NTSB National Transportation Safety Board
PHMSA Pipeline and Hazardous Materials Safety Administration
RITA Research and Innovative Technology Administration
SAE Society of Automotive Engineers
SAFETEA-LU Safe, Accountable, Flexible, Efficient Transportation Equity Act:
A Legacy for Users (2005)
TCRP Transit Cooperative Research Program
TDC Transit Development Corporation
TEA-21 Transportation Equity Act for the 21st Century (1998)
TRB Transportation Research Board
TSA Transportation Security Administration
U.S. DOT United States Department of Transportation

ADDRESS SERVICE REQUESTED
Washington, DC 20001
500 Fifth Street, NW
ISBN 978-0-309-67373-0
90000
9 780309 673730

The National Academies Press: Proposed Modification To AASHTO Cross-Frame Analysis and Design (2021)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The National Academies Press: Proposed Modification To AASHTO Cross-Frame Analysis and Design (2021)

Uploaded by

Copyright:

Available Formats

THE NATIONAL ACADEMIES PRESS

This PDF is available at http://nap.edu/26074 SHARE

Proposed Modification to AASHTO Cross-Frame Analysis and

156 pages | 8.5 x 11 | PAPERBACK

National Academies of Sciences, Engineering, and Medicine 2021. Proposed

– Access to free PDF downloads of thousands of scientiﬁc reports

Copyright © National Academy of Sciences. All rights reserved.

NCHRP RESEARCH REPORT 962

Copyright National Academy of Sciences. All rights reserved.

NATIONAL COOPERATIVE HIGHWAY NCHRP RESEARCH REPORT 962

Published research reports of the

NATIONAL COOPERATIVE HIGHWAY RESEARCH PROGRAM

Transportation Research Board

and can be ordered through the Internet by going to

Copyright National Academy of Sciences. All rights reserved.

Learn more about the Transportation Research Board at www.TRB.org.

Copyright National Academy of Sciences. All rights reserved.

COOPERATIVE RESEARCH PROGRAMS

CRP STAFF FOR NCHRP RESEARCH REPORT 962

NCHRP PROJECT 12-113 PANEL

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

12 Chapter 2 Research Approach

Copyright National Academy of Sciences. All rights reserved.

60 Chapter 3 Findings and Applications

134 Chapter 4 Conclusions and Suggested Research

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

2 Proposed Modification to AASHTO Cross-Frame Analysis and Design

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

1.1 Problem Statement and Current Knowledge

Copyright National Academy of Sciences. All rights reserved.

1.1.1 Best Practices for Cross-Frame Layout and Detailing

Copyright National Academy of Sciences. All rights reserved.

6 Proposed Modification to AASHTO Cross-Frame Analysis and Design

1.1.2 Cross-Frame Fatigue Behavior

Copyright National Academy of Sciences. All rights reserved.

1.1.3 Cross-Frame Analysis

1.1.4 Cross-Frame Stability Bracing Requirements

Copyright National Academy of Sciences. All rights reserved.

8 Proposed Modification to AASHTO Cross-Frame Analysis and Design

1.2 Research Objectives

Copyright National Academy of Sciences. All rights reserved.

c. Additional guidance on how to evaluate the influence of end connections on cross-frame

Copyright National Academy of Sciences. All rights reserved.

10 Proposed Modification to AASHTO Cross-Frame Analysis and Design

1.3 Scope of Study

1.4 Organization of Report

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

Research Approach 13

Phase Perform background

Phase Conduct field experiments

Phase Validate FEA modeling

Phase Conduct parametric FEA

Phase Develop specification and

Phase Finalize report and prepare

Figure 2-1. General flow of work related to NCHRP Project 12-113

Copyright National Academy of Sciences. All rights reserved.

14 Proposed Modification to AASHTO Cross-Frame Analysis and Design