Citation
Non-destructive evaluation of the 6th Avenue viaduct using strain gauges

Material Information

Title:
Non-destructive evaluation of the 6th Avenue viaduct using strain gauges
Creator:
Allen, Benjamin J
Publication Date:
Language:
English
Physical Description:
xii, 246 leaves : illustrations ; 28 cm

Subjects

Subjects / Keywords:
Thermal stresses ( lcsh )
Strain gages ( lcsh )
Concrete construction -- Bearing pads ( lcsh )
Concrete construction -- Bearing pads ( fast )
Strain gages ( fast )
Thermal stresses ( fast )
Sixth Avenue (Denver, Colo.) ( lcsh )
Colorado -- Denver ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 245-246).
General Note:
Department of Civil Engineering
Statement of Responsibility:
by Benjamin J. Allen.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
50727550 ( OCLC )
ocm50727550
Classification:
LD1190.E53 2002m .A54 ( lcc )

Full Text
NON-DESTRUCTIVE EVALUATION OF THE 6 AVENUE VIADUCT
USING STRAIN GAUGES
by
Benjamin J. Allen
B.S., University of Colorado at Denver, 1999
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment of the
requirements for the degree of
Master of Science
Civil Engineering


This thesis for the Master of Science
degree by
Benjamin John Allen
has been approved
by
Brian T. Brady
f-26- 02
Date
11


Allen, Benjamin John (M.S., Civil Engineering)
Non-Destructive Evaluation of the 6th Avenue Viaduct Using Strain Gauges
Thesis directed by Associate Professor Kevin L. Rens
ABSTRACT
This thesis presents the procedure and results of a strain gauge study
conducted on the Eastbound 6th Avenue Viaduct Bridge in Denver, Colorado. The 6th
Avenue Viaduct was modified in a 1998 rehabilitation making the entire
superstructure continuous. New sliding bearing pads failed to properly accommodate
expansion and contraction of the bridge superstructure due to temperature changes.
Visible damage due to contraction of the superstructure in the winter of 1999-2000
prompted a study of substructure elements to determine the short-term safety and long
term viability of the bridge. Substructure elements were instrumented with a total of
62 strain gauges and 2 concrete crack gauges. Data was taken for a period of three
months. Results of the study suggest that the bridge is safe for short-term use, but
that the long-term viability of the structure has been compromised. Repair or
replacement of substructure elements is recommended, along with frequent
monitoring during the interim period.
in


This abstract accurately represents the content of the candidates thesis. I recommend
its publication.
Signed
Kevin L. Rens


DEDICATION
I dedicate this thesis to my teacher and friend Dr. Kevin Rens. Dr. Rens exemplifies
the highest standards of excellence in teaching, professionalism, caring, and
dedication that all faculty at the University of Colorado should strive to achieve.
Without the support, understanding, and patience of Dr. Rens this thesis would not
have been possible.
v


ACKNOWLEDGEMENT
I would like to acknowledge the following people who assisted in the work required
for the study upon which this thesis is based and who supported the process.
University of Colorado at Denver
> Kevin Rens
> Myron LaCome
V Mike Gogel
> Deborah Steckley
> Yvonne Neiman
The City and County of Denver
> Terry Gruber
> Edward C. Maez Jr.
y William Melton
The URS Corporation
> Chengyu Li
vi


CONTENTS
Figures....................................................................x
Tables...................................................................xii
Chapter
1. Introduction............................................................1
1.1 Background.............................................................1
1.2 Scope.................................................................12
1.3 Procedure.............................................................14
2. Previous Research......................................................15
2.1 Introduction..........................................................15
2.2 Purpose of Field Bridge Testing......................................16
2.3 Applications and Methodology of Bridge Testing.......................18
2.4 Testing and Analysis of Bridges for Thermal Effects..................21
3. Strain Measurement Study...............................................22
3.1 Introduction..........................................................22
3.2 Equipment and Installation...........................................23
3.3 Strain Gauge Layout..................................................24
3.4 Linear Measurement Layout............................................31
vii


3.5 Crack Gauges for Pier 17.............................................35
3.6 Data Collection Procedure...........................................38
4. General Overview of Data Collected....................................41
4.1 Types of Data Collected..............................................41
4.2 Strain Data.........................................................43
4.3 Plumb Bob Data......................................................45
4.4 Abutment Measurement Data...........................................47
5. Steel Pier Analysis...................................................51
5.1 Introduction and Summary.............................................51
5.2 Data Analysis.......................................................52
5.3 Strain Results and Trends Observed..................................53
5.4 Determination of Short-Term Pier Safety.............................55
5.5 Determination of Long-Term Pier Safety..............................60
6. Pier 17 Analysis......................................................63
6.1 Introduction and Summary.............................................63
6.2 Data Analysis.......................................................63
6.3 Determination of Pier Safety........................................64
7. Summary and Conclusion................................................68
7.1 Summary of Findings..................................................68
7.2 Recommendations for Future Study.....................................69
viii


7.3 Future Corrective Repairs
70
Appendix
A. Typical Views of the 6th Avenue Viaduct............................72
B. Strain Gauge User Manual...........................................84
C. Calibration of Concrete Crack Gauges..............................137
D. Weekly Monitoring Strain Data.....................................141
E. 24-Hour Monitoring Strain Data....................................151
F. 34-Hour Monitoring Strain Data....................................163
G. Strain vs. Temperature Plots......................................181
H. Strain vs. Temperature Plots of Continuous Monitoring Sessions,
Columns 3 North and 16 North.......................................212
I. Pier 17 Crack Width vs. Temperature Plots.........................221
J. Pier 17 Crack Inventory............................................225
K. Linear Bridge Movements...........................................236
Bibliography..........................................................245
IX


FIGURES
Figures
1.1 Westbound 6th Avenue Viaduct...........................................2
1.2 View of Eastbound 6th Avenue Viaduct...................................3
1.3 Detail showing an elevation view of pin and hanger joint from original
plan set.............................................................5
1.4 Typical retrofit hinge configuration with plates welded across the joint
under the hanger bar, as well as on the top and bottom flange..........8
1.5 Sliding bearing pads at east abutment (top) and west abutment (bottom).9
1.6 Flexural cracking on Pier 17........................................11
3.1 Instrumented piers on the 6th Avenue Viaduct........................26
3.2 Typical placement of strain gauges..................................28
3.3 Strain gauge and protective covering on corroded section
at the base of Pier 16, south column................................29
3.4 Typical strain gauge layout on Pier 8 column.......................30
3.5 West abutment linear measurement location at barrier wall
expansion location..................................................32
3.6 Typical linear measurement location between end of girder and
bridge seat back wall at east abutment..............................33
x


3.7 Linear measurement using a plumb bob..................................34
3.8 Drawing of a concrete crack gauge.....................................36
3.9 Crack gauge and protective box mounted on Pier 17....................37
3.10 Close-up view of crack gauge mounted across a typical flexural
crack on Pier 17...................................................38
4.1 Typical Field Data..................................................42
4.2 Strain vs. Temperature plot for 24-hour data.......................44
4.3 Strain vs. Temperature plot for 34-hour data.......................45
4.4 Movement plots of plumb bobs for Pier 13, 24-hour monitoring session.46
4.5 24-hour linear measurement data for west bridge abutment...........48
4.6 Distance vs. Temperature plot for all data points, west abutment...49
5.1 Strain vs. Temperature plot for Pier 16 N, gauge 6..................53
6.1 Crack Width vs. Temperature, Pier 17................................64
xi


TABLES
Table
4.1 Linear displacement and temperature ranges...........................50
5.1 Force trends in strain gauges........................................54
5.2 Stress ranges (from strain measurements) and compressive stresses....55
5.3 Estimated loads used for calculation of axial stress.................57
5.4 Bending, calculated axial, and combined stresses with comparison
to yield stress......................................................57
5.5 Results of AASHTO Equations 10-42 and 10-43 for combined axial
compression and bending stresses.....................................59
5.6 Theoretical bending, calculated axial, and total stresses with
comparison to yield stress...........................................61
5.7 Results of AASHTO Equations 10-42 and 10-43 for combined axial
compression and bending stresses for theoretical bending stresses....61
6.1 Calculated steel stresses based on crack width range and absolute
maximum crack width..................................................65
xii


1. Introduction
1.1 Background
The 6th Avenue Viaduct Complex consists of two separate bridges crossing
Union Pacific Railroad and Regional Transportation District (RTD) Light Rail
property in central Denver, Colorado. On the west end of the viaduct is US Highway
6, an important regional highway in the Denver metro area, connecting the western
suburbs and Interstate 70 to Interstate 25 and major areas of commerce and
employment in central Denver. The east side the viaduct connects to 6th Avenue, a
major arterial street spanning most of the City of Denver.
Each bridge of the 6th Avenue Viaduct carries traffic in one direction. The
westbound structure shown Figure 1.1 was built in 1958. The substructure is
constructed of 18 single, round reinforced concrete piers with steel pier caps. The
superstructure consists of six wide-flange steel girder lines with a composite concrete
deck. The eastbound bridge shown in Figure 1.2 was built between 1963 and 1965,
with an additional girder line and traffic lane added in 1973. The eastbound bridge
has a steel substructure and eight wide flange girders with a composite concrete deck.
1


Figure 1.1
Westbound 6th
Avenue Viaduct.
Above: view from
Union Pacific rail
yard. Below:
Typical concrete
pier with steel pier
cap.
2


Figure 1.2 View of Eastbound 6th Avenue Viaduct. Above:
Looking east from Osage St. Below: Pier 15 looking east.
3


Whiles the superstructure of the eastbound bridge is very similar to the
westbound bridge, the substructure is quite different. Due to similarities in the
superstructures, both bridges received similar treatment in a 1997 rehabilitation
project. However, differences in the substructures precipitated a much different
response in the structural behavior by each bridge. The eastbound 6th Avenue
Viaduct is the focus of this study.
The eastbound 6th Avenue Viaduct originally utilized system of simple,
overhanging, and suspended spans for the superstructure. More specifically, in nine
locations along the bridge, girders would span between two piers with overhangs on
each end. In between adjacent overhangs, pins and hanger bars held a suspended
span in place. Figure 1.3 shows a typical pin and hanger connection on the
unmodified structure. Additional views of the structure can be found in Appendix A.
4


Figure 1.3 Detail showing an elevation view of pin and hanger joint from
original plan set (details can be found in Appendix A).
The hinge joint at each pin transferred shear forces, but no moment. Each
hinge joint was accompanied by an expansion joint in the bridge deck to
accommodate differential movement between the separate sections of the bridge, as
well as thermal expansion and contraction in the superstructure. In its original
configuration, the eastbound bridge consisted of 9 suspended spans, and 17 simple
spans with overhangs. Accordingly, there were 17 hinge points in the superstructure
of the bridge.
5


The eastbound 6th Avenue Viaduct is 1,430 feet in length and is supported by
a series of 17 piers. The bridge piers are numbered consecutively from east to west.
Piers 1 through 16 are steel piers, built up from wide flange sections, consisting of
either straight or A-frame piers. Straight steel piers consist of three separate wide
flange columns welded to a wide flange pier cap. A-frame piers are constructed of
three groups of wide flange sections that are welded in an A shape. Like the
straight piers, the A-frame piers have a pier cap consisting of a wide flange section
welded at the top. Unlike the others, Pier 17 is constructed of concrete. The pier is
constructed of five rectilinear columns and a rectangular concrete pier cap of varying
height. Eight wide flange girders of varying size bear on the top of each pier cap.
The concrete bridge deck is integral with the girders via shear studs.
As the metro area has grown over the last three decades, use of the 6th Avenue
Viaduct has increased. In 1998 the eastbound bridge carried an average of 31,630
vehicles per day, while the westbound bridge carried an average of 23,870 vehicles
per day. Unfortunately, the original structural system of the bridges did not
adequately accommodate increased loads from larger, heavier trucks, and general
increased use of the bridge. Most notably, the cantilevered and suspended portions of
bridge girders allowed for an unacceptable level of live load deflection at several
points on the bridge. The level of deflection created a bouncy feel that was both
noticeable and disconcerting to motorists on the bridge. Concerns about
serviceability, general deterioration of the superstructure, and safety concerns
6


regarding the pin and hanger bar connections prompted the City of Denver to initiate
a rehabilitation of both bridges that was completed in 1998.
During the rehabilitation the steel pins and hanger bars were removed at the
hinge locations, and steel plates were welded to the ends of the cantilever beams. A
typical hinge retrofit configuration is shown in Figure 1.4. This modification created
a continuous superstructure in each bridge that would help to reduce live load
deflections. In order to accommodate the thermal expansion and contraction of the
superstructure, the original rocker-type bearings were removed at the abutments and
replaced with elastomeric bearing pads with polytetrafluroetheylene (PTFE) and
stainless steel sliding surfaces as shown in Figure 1.5. Similar bearing pads were
placed under the girders at each pier, with the exception of Pier 8 on the eastbound
bridge and Pier 11 on the westbound bridge, where each structure was fixed
respectively. Expansion joints in each bridge deck, at locations corresponding to
previous hinge locations were also removed. New concrete deck was poured on these
locations to fill the voids and create a continuous deck surface over each entire span
of each bridge.
7


Figure 1.4 Typical retrofit hinge configuration with plates welded
across the joint under the hanger bar, as well as on the top and bottom
flange
8


Figure 1.5 Sliding bearing pads at east abutment (top) and west
abutment (bottom)
9


As the rehabilitation approached completion in late 1997, a project inspector
noticed horizontal cracks on the west face of Pier 17, the concrete pier on the
eastbound structure (see Figure 1.6 and Appendix J). The inspector alerted project
engineers who then inspected the pier and bearing devices. The engineers observed
that the cracking was consistent with flexure of the pier in the direction of the
longitudinal axis of the bridge. Upon inspection of the sliding bearing devices, it was
noted that the stainless steel/Teflon interface had not slipped since installation,
although the temperature had fluctuated enough to contract the structure substantially.
They concluded that the malfunctioning bearing pads on Pier 17 were causing a
temperature induced lateral load leading to flexural cracking on the west face of the
column.
10


Figure 1.6 Flexural cracking on Pier
17. Above: typical cracking on the
north face of a column. Below: detail
of a crack
11


Fearing that other bearing pads may be malfunctioning as well, engineers
inspected the bearings on other piers of the bridge and found that they did not appear
to sliding either. It was further concluded that additional piers might have been
subjected to the same temperature induced lateral load. The fact that Pier 17 was
suffering from severe flexural cracking provided indisputable evidence that the pier
was being loaded in a manner not consistent with its design. All of the other piers on
the structure were constructed of steel, giving them additional flexibility that did not
provide such obvious visual clues of distress. So while the cracking in Pier 17 was
problematic, possible overstress conditions in the steel piers, the long term effects of
malfunctioning bearings on the structure, and the chance of structural failure all
prompted the need for further investigation. Empirical data was necessary in order to
evaluate the extent and time frame of repairs to the behavior of the bridge.
1.2 Scope
Engineers with the City concluded that it was necessary to quantify the current
level of stress being carried by the piers in order to assess both the short and long
term viability if the structure, the damage caused by the functional problems of the
rehabilitation design, and provide data for a repair effort. It was concluded that all of
these questions could be answered readily if strain gauges were used to determine the
differential strain at the base of several piers over a period time during the winter of
1999-2000. The change in strain could be used to calculate the magnitude of stress
12


changes, the relation of stress changes to temperature, and the amount of movement
at the top of the pier. Furthermore, strain gauges are a proven technology and
relatively simple and inexpensive to operate.
The scope of this thesis is to:
1. Research the application of strain gauges to the problem
2. Review previous research and literature on bridge testing and in particular
testing with strain gauges
3. Develop a monitoring program to test the behavior of the Eastbound 6th
Avenue Viaduct
4. Design and fabricate crack gauges to monitor the flexural behavior of Pier 17
5. Instrument the viaduct as needed and measure strains in several columns and
linear displacements in the superstructure over both long and short-term
temperature cycles from January to April 2000
6. Determine the approximate level of stress in the bridge piers
7. Evaluate the short and long-term safety of the bridge based on the data
gathered
13


1.3 Procedure
The study began by placing a series of 64 strain gauges and appropriate
weatherproofing material on 11 columns of 6 piers of the viaduct. Four locations
were marked at the bridge abutments for taking linear measurements, and eleven
plumb bobs were hung from the pier caps at the locations of the strain gauges for
additional linear measurement. Ideally, the linear measurements would be used for
correlation purposes to help validate the strain gauge data.
Data was taken from late January through early April in two ways: weekly
data and long period data. Weekly data consisting set of one set of strain readings,
was gathered beginning at 7:00 AM for nine weeks. Two sets of long period data
lasting 24 and 34 hours were taken with, strain readings taken approximately every
two hours over the entire 24 or 34 hour period. Linear measurements, consisting of
distance measurement at each abutment and measurement of the plumb bob location
were taken in conjunction with the strain readings.
14


2. Previous Research
2.1 Introduction
The majority of bridge infrastructure in the United States was built during two
periods of substantial construction: the Depression Era of the 1930s and the
Interstate construction boom of the 1950s and 60s. The American Association of
State Highway and Transportation Officials (AASHTO) estimates that approximately
90,000 bridges built in the 1930s are reaching the end of their useful life and will
require replacement. Unlike these structures, about 223,000 bridges built between
1956 and 1975 will require major rehabilitation in the coming decade (AASHTO
Guidelines for Bridge Management Systems, 1993). As state and local agencies are
forced to maintain bridge infrastructure with increasingly smaller budgets, adequately
addressing the problem of aging bridge infrastructure has become more difficult
(Hughes et al, 1996).
To help deal with the problem, a number of methods have been developed in
recent years to help engineers address the problem of maintaining a functional bridge
network while limiting the number of bridge closures or weight restrictions. Bridge
management systems and deterioration modeling methodologies have emerged as
useful methods of predicting needs, prioritizing projects, and directing funding.
Unfortunately, these methods cannot provide empirical data on the actual condition
15


and behavior of deteriorated and distressed bridge infrastructure. Often such data are
required if critical decisions regarding safety, prioritization, and feasibility can be
made. As a result, field bridge testing has become increasingly popular as a tool for
engineers to assess the performance of bridges (Saraf and Nowak, 1998; Gogel,
2001). Such testing provides valuable information for repairs and rehabilitation
strategies on both case specific and network level depending upon the application
method. Indeed, many agencies are adopting network-wide testing programs for
evaluation of bridge infrastructure (Moses et al, 1994; DeWolf et al 1998).
In addition to addressing specific needs within an agency, increased use of
bridge testing under service level conditions has led to a better understating of actual
structural behavior. Over time this knowledge will enable engineers to more
accurately predict bridge behavior and inevitably lead to changes in building codes.
2.2 Purpose of Field Bridge Testing
As with all structural engineering, the design of bridges incorporates a set of
assumptions. These assumptions are often prescribed by building codes or result
from the application of theory to practice. For the purposes of simplicity and safety,
both design and evaluation of bridges are carried out using assumptions regarding
bridge behavior that are often conservative (Saraf and Nowak, 1998; Sartor et al,
1999). Assumptions and models often fall short of accurately or adequately
predicting actual bridge behavior. Prescribed design loading conditions are almost
16


never encountered. Furthermore, it is not always possible for the most experienced
bridge engineer to consider and account for all possible variables in a complicated
design (Sartor et al, 1999).
As bridge infrastructure deteriorates, many assumptions regarding actual
material properties or assumed behaviors change (Moses et al, 1994, Sartor et al,
1999). Biennial inspection for safety purposes relies heavily on visual techniques.
Visual inspections often indicate a need for further testing that can nicely augment
visual inspection data and aid in analysis. In extreme cases, visual inspections can
miss critical elements and can be entirely inadequate for some purposes (DeWolf et
al, 1998; Mohammadi et al, 1998).
A growing need for data on the in-situ performance of bridges is currently
being met by growing interest and practice in the area of field bridge testing.
Currently, field testing of bridges serves two principal purposes:
1. Verification of models and assumptions used in bridge design
2. Evaluation of existing infrastructure
Testing for both purposes has become more popular as improved technology and
declining equipment costs in recent years have made bridge testing more feasible than
ever.
The advantage of carrying out bridge testing for both purposes was stated
most effectively by Bakht and Jaeger in their paper Bridge Testing A Surprise Every
Time.
17


There is no better way for a bridge engineer to
understand the shortcomings of the mathematical
models used for the design or evaluation of bridges than
to investigate the behavior of bridges through field-
testing.
2.3 Applications and Methodology of Bridge Testing
As bridge testing has increased in recent years, so have the number of
applications of testing (Bakht and Jaeger, 1990; DeWolf et al, 1998; Chajes et al,
1997, Boothby and Craig, 1997; Mohammadi et al, 1998; Saraf and Nowak, 1998;
Sartor et al, 1999; McElwain and Laman, 2000; Prine, 1996; Moorty and Roeder,
1992; Moses et al, 1994; Farhey, 2000). Applications of testing methods and
technology include:
1. Estimation of fatigue life in existing steel bridge girders under current
traffic loads
2. Verification of design assumptions and guidelines for curved steel
girder bridges
3. Verification of bridge rating methods
4. Verification of calculated bridge load capacity for posted bridges
5. Assessment of deterioration in existing historic truss bridges
18


6. Assessment of the need for replacement, rehabilitation, or repair of
existing infrastructure
7. Determination of thermal effects in bridges and verification of code
guidelines for thermal effects
8. Verification and quantification of vibration and serviceability concerns
9. Monitoring of safety concerns in critical, deteriorated bridge elements
10. Determination of the extent effect of non-structural elements to overall
bridge behavior and capacity
11. Determination of the level of composite action between non-composite
members in slab on girder steel bridges
Each of these items can be categorized for the purposes of either verification
or evaluation. However, the scope and duration of bridge testing is generally tailored
to the specific goals of the study. In some cases a bridge is tested in a single day,
while other studies may involve permanent instrumentation or retesting carried out at
specified intervals. Often specific elements of a bridge are targeted for
instrumentation because initial analysis or visual inspection suggests areas of
concern.
A more rigorous, systematic, and standardized approach to bridge testing has
been adopted in the past in Switzerland and Canada (Moses et al, 1994; Bakht and
Jaeger, 1990). In Switzerland, a program of testing new bridges upon completion of
19


construction to verify behavior assumptions used in design has been used for several
years. Such tests are conducted by proof loading bridges to a substantial portion of
total design loads and measuring deflections. All major bridges are tested before
opening for traffic. The Ontario Ministry of Transportation conducted an extensive
program of testing over 225 older, deteriorated bridges for diagnostic purposes in the
late 1980s. Such extensive studies have proven useful in generating a good base of
knowledge in a relatively new area of study.
While some of the specific methodology being used for various applications
of field testing is relatively new, much of the technology being used in these
applications has been in use for several decades. In particular, bridge test methods
using strain gauges in conjunction with new technology, have become more
prominent in recent years. The ability to measure actual field stresses in bridge
elements is unmatched by any other type of testing technology and is the principal
advantage of using strain gauges.
Literature reviewed for this study indicates that the vast majority of tests rely
on stain gauges as the principal test method. Other methods, such as displacement
measurement, are often used in conjunction with strain gauges to correlate, verify, or
augment data obtained in testing.
20


2.4 Testing and Analysis of Bridges for Thermal Effects
Though design guidelines address the issue of thermal movements and
stresses in bridge structures, actual design can be extremely complicated. Small
miscalculations or oversights can introduce large forces into elements that may not be
capable of adequately supporting such loads. Consequently, the accurate prediction
of potential expansion and contraction in bridges is an important element of proper
design (Moorty and Roeder, 1992).
The American Association of Highway and Transportation Officials
(AASHTO) required that the effects of thermal expansion be properly considered in
design. AASHTO also specifies a range of temperature that must be accommodated
based upon geographic location. These methods work reasonably well for calculating
the longitudinal expansion, but fail to properly consider more complex behaviors.
Studies have indicated that support conditions and geometry heavily influence the
thermal behavior of bridges, and such factors are especially pronounced in skewed
bridges (Moorty and Roeder, 1992).
Currently, very little empirical data regarding bridge temperature and thermal
movement exists for bridges in the United States. Analysis indicates that curved or
skewed bridges are more susceptible to damage from un-anticipated thermal loads.
Further study in his area is needed to help augment code requirements and provide
greater understanding of current problems (Moorty and Roeder, 1992).
21


3. Strain Measurement Study
3.1 Introduction
The strain measurement study was conducted in order to provide empirical
data that would perform three primary functions. First, the data would serve to verify
conclusions that the bridge piers were being subjected to temperature induced lateral
load cycles resulting from the malfunctioning bearing pads. Secondly, the data would
enable the quantification of the level of stress being carried by the bridge piers to
determine the short-term safety of the bridge. Finally, the empirical data would allow
for a broad scale condition assessment of the bridge to determine the extent of the
damage incurred and the scope of a long-term solution for the structural problems.
The majority of the modifications made during the rehabilitation of the bridge
were made during the summer of 1997. During this period the expansion joints in the
deck were removed, and the hinge points in the superstructure were fixed by welding
steel plates between adjoining beams as shown in Chapter 1. As the joints were
fixed, the existing bearing pads were removed and replaced with sliding bearings
which were supposed to accommodate the thermal expansion and contraction of the
superstructure. The average daily temperatures during the period when this work was
accomplished were relatively high, and the structure was more or less fully expanded.
Consequently, the extreme cold temperatures of the winter months would produce the
22


greatest differential temperature strain. For this reason the strain gauge study was
conducted during the months of January through April.
3.2 Equipment and Installation
The strain study used equipment and associated materials manufactured by the
Measurements Group Inc.1, and were selected based upon the specific needs of the
study. Several factors were considered:
1. The majority of the work involving the strain gauges, including
mounting was to be conducted in the field.
2. Strain gauges were to be mounted on the bridge during the months of
November, December, and January when the weather conditions were
unfavorable for epoxy curing.
3. Strain reading equipment needed to accommodate data collection from
up to seven gauges simultaneously.
4. All equipment needed to be easily portable to allow for reading strains
on separate columns over the 14 mile length of the viaduct.
5. Additional weatherproofing protection for the gauges was necessary in
order to prevent moisture penetration and maintain functionality over a
period of several months.
1 Measurements Group Incorporated, Raleigh, North Carolina 27611
23


The strain gauges used for the study were all Measurements Group type CEA-
06-500UW-120 gauges. These gauges were selected for their ability to accommodate
a fairly large strain range and their larger size, which would allow for easier field
handling and installation. The gauges were attached using M-Bond 200 epoxy. The
epoxy required minimal curing and clamping time under lower temperatures, but
provided a level of durability that would accommodate the length of the study. Once
mounted, the gauges were covered using the Measurements Group M-Coat F Kit,
which includes a butyl rubber sealant, rubber padding, and aluminum foil tape.
Silicone caulk was also used to provide an additional level of protection from
moisture. To read strains, the Measurements Group P-3500 Portable Strain Indicator
was selected, and was used in conjunction with the SB-20 Switch and Balance Unit.
The SB-20 unit contains 20 separate channels and balance knobs. Use of the SB-20
allowed for continuity of data by facilitating quick, successive reading of strains from
up to seven gauges at once. For more detailed descriptions of the equipment used and
the installation process, see Appendix B.
3.3 Strain Gauge Layout
The bridge is fixed at Pier 8, corresponding roughly to the midpoint of the
span. The lengths of the bridge piers are greater on the west side of the bridge and
were of greater concern because of the larger bending moments introduced by the
24


contraction of the superstructure. Accordingly, the majority of the gauges were
placed on piers west of the midpoint.
Columns were selected for gauge placement because the specific column or
pier was of particular concern, or the location would provide a basis for determining
the overall behavior of the structure. Columns were selected for gauge placement as
follows:
Piers 3 and 13 were selected as representative straight steel piers on
either side of the Pier 8. Both piers are approximately the same
distance from the fixed point at Pier 8.
Pier 8 was selected because it is the point at which the superstructure
is fixed. In theory, this pier would be subjected to little or no lateral
loads as the bridge expanded and contracted during temperature
cycles. If the entire structure was being induced to shift globally
during temperature cycles, monitoring of Pier 8 would help provide
indications of such behavior.
Pier 16 was selected because significant section loss from corrosion
was discovered at the base of the north and south columns of the pier.
Pier 17 was monitored because of the severe flexural cracking on the
west face of the columns.
25


In all, a total of ten separate wide flange sections were instrumented. The locations of
the instrumented columns are shown in Figure 3.1
26


For all steel piers, strain gauges were placed on both the north and south
columns. Gauges were placed approximately 1-2 feet above the top of the concrete
encasement at the bottom of each column. At this location, the gauges would provide
data very near to the point of maximum bending strain, and be in a practical position
for installation and protection from the elements.
Six gauges were attached to each wide flange column with the exception of
the columns on Pier 16 where an additional gauge was attached to each column at the
point of section loss. For each straight column, four gauges were attached to the tips
of the flanges and two were placed on the top of each flange at the point where the
web connects as seen in Figure 3.2. On Pier 16 an additional gauge was attached to
each column at the point of maximum section loss as shown in Figure 3.3. For Pier 8,
gauges were placed both sides of the A-frame in the same manner as on the straight
piers. However, the gauges were placed at variable heights from the concrete
pedestal due to the inclination of the individual sections. This arrangement is shown
in Figure 3.4. All gauges on Pier 8 were placed to measure strain in the direction
parallel to the longitudinal axis of the column.
Uniform wire lengths of four feet were attached to each gauge to provide
consistent data and facilitate ease of collection. Gauges were numbered in a
consistent manner on each column. The wire leads were numbered accordingly and
bundled at a common point.
27


Strain Gauge
/
I
i
Typical Column Cross-Section
Figure 3.2 Typical placement of strain gauges.
Above: Placement shown on cross-section
Below: Photograph of gauges in the field
28


Figure 3.3 Strain gauge and protective covering on corroded
section at base of Pier 16, south column
29


30


3.4 Linear Measurement Layout
As a means of confirming the accuracy of the strain gauge readings and
correlating the actual physical movement of the bridge with the recorded strains,
linear measurements were taken in conjunction with strain readings. One location
was marked at the west abutment of the bridge at an expansion joint of the southern
barrier wall. Three locations were marked between the ends of girders 3, 5, and 7 and
the back wall of the bridge seat at the east abutment. These measurement locations
are shown in Figure 3.5 and 3.6.
31


.'f
Figure 3.5 West abutment linear measurement location at
barrier wall expansion location
32


Figure 3.6 Typical linear measurement location
between end of girder and bridge seat back wall at east
abutment
33


In addition to the measurements taken at the abutments, eleven plumb bobs
were also installed on bridge piers with strain gauges. The plumb bobs were attached
to the north and south ends of the pier caps, with two control monuments placed in
the ground on the east and west side of the plumb bob. The intent of the plumb bobs
was to provide a linear measure of any lateral movement of the pier cap. A
photograph of plumb bobs measurement is shown in Figure 3.7.
Figure 3.7 Linear measurement using a
plumb bob
34


3.5 Crack Gauges for Pier 17
As the only concrete pier on the bridge, Pier 17 required special consideration.
Differences in materials required different methodology. While an accurate means of
measuring the behavior of the pier was sought, some level of continuity in the study
was also desired. It was determined that the best means of measuring the flexural
behavior of the pier was through measurement of crack widths on the tension face of
the columns. To accomplish this, special crack gauges were fabricated and mounted
across cracks on two columns of Pier 17.
The crack gauges consisted of a spring steel frame and an internal steel spring
on which a strain gauge was mounted. Figure 3.8 shows a drawing of the crack
gauges. As the frame of the gauge is pushed or pulled, the tension strain on the face
of the internal spring varies and the strain gauge measures this difference. After these
crack gauges were fabricated, they were calibrated by pushing and pulling a measured
amount and recording the resulting strain. The calibration produced an equation of a
linear relationship between strain and displacement. These equations were used to
determine the change in the width of the crack opening based on the strain read.
More detail on the crack gauge calibration procedure and results can be found in
Appendix C.
35


Frame made of 3/32" spring steel
os
Tension arc made of '64" spring steel
Strain gage mounted at top of arc
Figure 3.8 Drawing of concrete crack gauge


Once fabricated and calibrated, both gauges were mounted across flexural
cracks on Pier 17 using epoxy. The crack gauges were also covered with metal boxes
to protect them from damage. Photographs of the crack gauges mounted to the pier
can be seen in Figure 3.9 and 3.10.
Figure 3.9 Crack gauge and protective box mounted on
Pier 17.
37


Figure 3.10 Close-up view of crack gauge mounted across a typical
flexural crack on Pier 17.
3.6 Data Collection Procedure
Data was taken over a period of over three months. For the sake of
consistency, data was always taken in order of west to east. Plumb bob
measurements for each pier were taken at the same time that corresponding strains
were taken. A thermometer was used to record the ambient air temperature at the
same time strains were measured.
38


The following standard operating procedure was employed for data collection:
1. Linear distance measurement was taken at the barrier wall at the west
abutment
2. The strains from the strain gauges and the distance measurement for the
plumb bob on each pier were taken in the following order:
17 North
17 South
16 North
16 South
13 North
13 South
8 Northwest
8 Southwest
8 Northeast
8 Southeast
3 North
3 South
3. Linear distance measurements were taken for Girders 3, 5, and 7 at the east
abutment
39


Completion of the data collection process required 1 V2 to 2 hours. Consequently, the
entire cycle of steps 1 through 3 was repeated approximately every 2 hours if the
monitoring session was carried out over an extended period of time.
40


4. General Overview of Data Collected
4.1 Types of Data Collected
A total of 37 separate sets of strain data were taken over the course of the
study. Three types of data were taken:
1. Weekly data
2. 24 hour monitoring data
3. 34 hour monitoring data
The three types of data represent two separate goals of study. Weekly data was taken
to monitor the status of the structure over a period of several weeks and several daily
temperature cycles. The purpose of the weekly data was to insure that repeated
expansion and contraction of the bridge superstructure was leading to significant
damage or plastic deformations of the pier columns. The two periods of extended
short-term monitoring were designed to record the behavior of the bridge over a
single temperature cycle. This data would verify that the bridge piers were indeed
being subjected to cyclic load cycles resulting form expansion and contraction of the
bridge superstructure. Typical data collected is presented in Figure 4.1
41


West 6th Avenue Viaduct Eastbound
24 Hour Monitoring Session Strain Report
w
|Date. 01/10/00
Gage Factor: 2.065 **
P-3500 Serial #: 0121053
Measured By: BJA/MDL/YN
Recorded By: CYL
Strain Gage 1 2 3 4 5 6 7
Pier-Column Temp Time Strain Strain Strain Strain Strain Strain Strain
17* 53.4 12:10 -163 -310 : 1 J J
16-S 475 12:45 -50 38 144 92 -13 -113 -423
16-N 443 12 26 -75 13 95 80 -1 -82 242
13-S 479 13:03 -13 29 74 35 -27 -64
13-N 446 12:55 -28 19 78 39 -10 -45
8-NW 443 13:45 -147 -123 -138 -151 -135 -132
8-NE 43.5 13:50 148 -135 -149 -140 -128 -126
8-SW 47.7 13 25 -48 -76 -113 -123 -105 81
8-SE 45 8 13:35 -149 -123 -108 -174 -165 25
3-S 45.5 14:05 90 15 -57 -75 -9 63
3-N 42.8 14:00 74 22 -52 35 -13 58
Plumb Bob Measurement East Abutment Measurement
Pier-Column Distance to West (in) Distance to East (in) Time Temp ("F) Distance to Girder 3 (in) Distance to Girder 5 (in) Distance to Girder 7 (in) Time
17-S (1) 13.8125 13.2500 12:10 53.4 6.9375 7.2500 6.5000 14:10
17-S (2) 17.6875 13.7500 12.10 53.4
17-N 12.1875 13.9375 12:20 46.9
16-S 8.5000 75625 12:35 47.5 West Abutment Measurement
16-N 9.1250 5.1250 12:25 44 3 Distance to
13-S 285000 32.5625 13:00 50.2 Barrier Time Temp (F)
13-N 41.4375 37.5625 12:55 45.5 (in)
8-S 69.8125 44 0000 13:20 47.7 6.9375 12 04 626
8-N 8.7500 16.5625 13:37 45.2
3-S 23.9375 25.0625 14 04 45.5
3-N 21.8125 24.0000 13:47 42 8
* Crack gages are numbered from South to North
* All gages have identical Gage Factor ( Type: CEA-06-500UW-120)
Note: All temperatures are reported in degrees Fahrenheit and all strains are reported in microstrains (10"6)
Field Notes:
None
Figure 4.1 Typical field data
42


4.2 Strain Data
A strain vs. temperature plot generated form the 24-hour monitoring is
presented in Figure 4.2. Originally the 24-hour monitoring session was intended to
record the behavior of the bridge over a full temperature cycle. As seen in Figure 4.2,
the structure did not complete a full, closed strain cycle during this period. In order to
capture a full cycle, a subsequent 34-hour session of short-term continuous
monitoring was undertaken shortly after the completion of the 24-hour cycle. Data
taken during this session produced a full cycle of strain data, as seen in Figure 4.3.
Plots of strain vs. temperature for the north columns of Piers 3 and 16 can be found in
Appendix H.
43


44


4.3 Plumb Bob Data
Plumb bob data proved to be of little value to the study because of the erratic
behavior the data indicate. The distance from the plumb bob tip to two surveyed
monuments was measured and recorded each time strain data was taken.
Trigonometric formulas were used to calculate an x and y coordinate based on the
recorded measurements. Plots of the plumb bob movements on Pier 13 during the 24
hour monitoring session are shown in Figure 4.4.
45


Figure 4.4 Movement plots of plumb bobs for Pier 13, 24 hour
monitoring session
46


As seen in Figure 4.4, the recorded movement of the plumb bobs are erratic
and do not exhibit any kind of linear behavior. Plots for the other plumb bobs
illustrate similar problems. A number of factors likely contributed to the poor data,
including:
1. Instability of the plumb bobs during measurement due to the height of the
pier cap, wind, and vibration from traffic on the bridge
2. Lack of precise boundary conditions for measurement
3. Variable distances between the plumb bobs and the monuments, ranging
from several inches to several feet
4. Use of a flexible tape measure
5. Human error
Problems with the data from the 24 hour session were duplicated in the 34 hour
session. Subsequent weekly monitoring only measured plumb bobs at Piers 16 and
17. Final analysis showed all plumb bob data to be of no value to the study.
4.4 Abutment Measurement Data
Data from linear measurements taken at the abutments were used to track the actual
expansion and contraction of the bridge superstructure. In the absence of useful
plumb bob data, these measurements were the only way to quantify of the actual
superstructure movement. For 24 and 34 hour monitoring sessions, charts of
47


Displacement vs. Temperature were created from the data. Figure 4.5 shows a typical
chart from the 24-hour session. All charts can be found in Appendix K.
Figure 4.5 24-hour linear measurement data for west bridge abutment
Plots of all data points taken during the study for the west abutment and
Girders 1-3 show a loose linear relationship between strain and temperature. Such a
relationship is illustrated in Figure 4.6. As can be seen in the plots, the bridge
contracted and expanded in agreement with temperature cycles.
48


More important to the study though, is the temperature range and actual vs.
theoretical displacement at the abutments. Table 4.1 lists the range for recorded
temperature and displacement, as well as the theoretical displacement for the recorded
temperature range. These values illustrate a marked discrepancy between the actual
and theoretical displacement at the west abutment, but a much closer agreement at the
east side of the bridge.
Figure 4.6 Distance vs. Temperature plot for all data points, west abutment
49


Table 4.1 Linear displacement and temperature ranges
Temperatures reported in degrees Fahrenheit. Displacements reported in
inches.
Location Min Temp Max Temp Temp Range Measured Displacement Theoretical Displacement Difference
West Abutment 18.8 62.6 43.8 1.3125 2.6648 1.3523
Girder 3 21.3 51.0 29.7 1.3750 1.5058 0.1308
Girder 5 21.3 51.0 29.7 1.2500 1.5058 0.2558
Girder 7 21.3 51.0 29.7 1.3125 1.5058 0.1933
Overall, the displacement data are important as they demonstrate the actual
contraction of the superstructure. In turn, it can be surmised that the tops bridge piers
are being displaced. However, the linear measurements concurrently demonstrate
some important unknowns with regard to the overall behavior of the bridge. These
unknowns include:
The amount of displacement at the top of the piers
The restraining force that the piers are exerting on the expansion of the
superstructure both overall, and on an individual level
Variability of recorded air temperatures at different points along the bridge
and probable discrepancies between air and structure temperature
These unknowns create problems with structural analysis based solely on linear data,
and underscore the importance of the strain data for determining the safety of the
bridge.
50


5. Steel Pier Analysis
5.1 Introduction and Summary
As ductile members, the steel bridge piers did not show visible signs of
distress following the modifications to the viaduct. While their ductility allowed for
greater distortions without damage, possible plastic deformations and critical
elements needed to be identified. Strain data were used to asses the steel piers in two
ways:
1. The current strains were used to determine the immediate safety of the
steel bridge piers
2. Behavior established by the strain data were used to assess the long-term
safety of the bridge
Analysis of the data suggests that Pier 16 requires replacement and that regular
monitoring should occur until repairs are made. Furthermore, Piers 3 and 13 do not
meet requirements of the Standard Specifications for Highway Bridges by American
Association of State Highway and Transportation Officials (AASHTO). These piers
will also have to be replaced to satisfy long-term safety issues.
51


5.2 Data Analysis
A plot of strain vs. temperature was produced for each gauge using all data
points collected during both weekly and extended monitoring sessions. As discussed
in section 4.2, cyclical behavior was observed in both the 24 and 34-hour sessions and
indicated that elastic behavior in the steel piers could be assumed. Therefore, a linear
relationship between temperature and strain was also assumed. Linear regression was
performed on each data set to produce a best-fit line and equation representing the
relationship between strain and temperature. Figure 5.1 shows the plot of strain vs.
temperature for Pier 16 North Gauge 6. Plots for all gauges can be found in
Appendix G.
52


Figure 5.1 Strain vs. Temperature plot for Pier 16 N, Gauge 6
5.3 Strain Results and Trends Observed
Overall, data collected confirmed presumptions regarding the behavior of the
bridge prior to study. Most importantly, the data confirm that the bridge piers were
bending inward toward Pier 8 with the expansion and contraction of the
superstructure. This behavior is illustrated by the Strain vs. Temperature plots which
indicate tension and compression forces on opposite side of the columns in Piers 3,
53


13, and 16. The data for these piers are consistent over the entire period of study.
The force trend for each gauge is summarized in Table 5.1.
Table 5.1 Force trends in strain gauges
C = Increase in compression for decrease in temperature
T = Increase in tension for decrease in temperature
Pier 3 3 8 8 8 8 13 13 16 16
Column N S NE SE NW sw N S N S
Gauge 1 C c C C C c T T T T
Gauge 2 T T C c c c C C C C
Gauge 3 T T c c c c C C C C
Gauge 4 T T T T c c C C C C
Gauge 5 T T C T c c C C C C
Gauge 6 C C C T c c T T T T
Gauge 7 C T
It should be noted that gauges in the tips of the flanges experienced a greater
change in strain than those on the side of the pier. This is consistent with their
locations relative to the neutral axis of bending. Further consistent with the tension
caused by the contraction of the bridge superstructure, most of the gauges on Pier 8
exhibit an increase in compression forces as temperature decreases, with the
exception of four gauges. Such behavior may be caused by the variable stiffness of
the different length members on Pier 8. In general, it appears that the columns of Pier
8 act as compression struts to resist the tension generated in the bridge superstructure.
54


5.4 Determination of Short-Term Pier Safety
Upon confirmation of bending in the steel piers, the safety of the piers was
evaluated. This evaluation included both the short-term or immediate safety, and the
long-term viability of the substructure.
Short-term safety was addressed first by determining the stress range and
maximum compressive stress for each column from the strain data. These results are
presented in Table 5.2. Alone, these results were not sufficient to allow for any
conclusive determinations. In order to determine the total compressive stress in the
columns, an estimation of the dead load of the structure and the live load carried by
the bridge was necessary.
Table 5.2 Stress ranges (from strain
measurements) and compressive stresses
Pier Stress Range Maximum Compressive
Stress
(psi) (psi)
3N 10,411 7,366
3S 8,932 6,699
8NE 2,407 5,539
8NW 2,436 5,655
8SE 5,046 7,453
8SW 5,046 7,569
13N 9,222 7,888
13S 6,873 5,075
16N 9,773 4,437
16S 15,863 15,254
55


Both dead and live load for the individual columns were calculated using the
tributary area of bridge deck carried by the individual piers. Values used for dead and
live load calculations are presented in Table 5.3. Finding the total load for the pier,
and then dividing the result by three (three columns per pier) determined a value for
the axial stress in each column. In the case of Pier 8, the total of dead and live loads
were resolved into the axial load in the legs. After the axial stresses were calculated,
they were then added to the maximum bending stress measured during the study to
produce a combined total stress for each pier. The combined stress was the stress
used for analysis and represents the total of the following:
Residual dead load stresses present prior to instrumentation of the
bridge
Live load stresses (To be conservative, all were assumed residual)
The maximum measured bending stress from field data
These combined stress values were initially compared to the yield strength of
the steel. The combined stress results along with a comparison to yield stress are
presented in Table 5.4.
56


Table 5.3 Estimated loads used for calculation of axial stress
Item Value
2" Asphalt Overlay 24 psf
7" Concrete Deck 87.5 psf
1.5" Haunch 18.75 plf
Type 4 Bridge Rail 465 plf
Type 4 Barrier Wall 460 plf
Steel Girders and Attachments 1435 plf of bridge
Miscellaneous 100 plf of bridge
Lane Live Load 640 plf/lane
Table 5.4 Bending, calculated axial, and combined stresses
with comparison to yield stress
Pier Bending Stress (from strain data) (psi) Calculated Axial Stress (psi) Combined Stress (psi) % of Yield Strength
3N 7,366 6,324 13,690 38
3S 6,699 6,324 13,023 36
8NE 5,539 3,131 8,670 24
8NW 5,655 3,131 8,786 24
8SE 7,453 3,131 10,584 29
8SW 7,569 3,131 10,700 30
13N 7,888 7,514 15,402 43
13S 5,075 7,514 12,589 35
16N 4,437 4,716 9,153 25
16S 15,254 4,716 19,970 55
57


Based on the results presented in Table 5.4, it can be presumed that the bridge
is safe for the immediate short-term. The combined axial and bending stresses total
less than the yield strength of 36,000 psi for all columns. Furthermore, all but two
columns have combined stresses less than 50% of the yield strength. However, the
combination of axial compression and bending in the piers carries added problems
with local yielding and buckling of the flange, as well as general column buckling.
AASHTO requires steel members subject to axial compression and bending to satisfy
two equations2:
L
K
+
^mfb
1-
L
e J
<1.0
(5-1)
and
fa h
0.472F; Fh
<1.0
(5-2)
where fa and/* for each column are equal to the values of Calculated Axial Stress and
Bending stress in Table 5.4 respectively.
Equations 5-1 and 5-2 (AASHTO Equations 10-42 and 10-43 respectively)
were used to evaluate the bridge piers using the field measured bending stresses and
the calculated axial stresses reported in Table 5.4. Results are presented in Table 5.5.
2 Standard Specifications for Highway Bridges, AASHTO, 1996
58


Pier 8 was not evaluated by these equations because it is not subject to substantial
bending stresses.
Table 5.5 Results of AASHTO Equations 10-42 and 10-43 for combined
axial compression and bending stresses
Pier Result AASHTO Eq. 10-42 Pass/Fail (Pass if <1.0) Eq. 10-42 Result AASHTO Eq. 10-43 Pass/Fail (Pass if <1.0) Eq. 10-43 Net Result
3N 0.75 Pass 0.74 Pass Pass
3S 0.72 Pass 0.71 Pass Pass
13N 2.80 Fail 0.84 Pass Fail
13S 2.10 Fail 0.70 Pass Fail
16N 0.82 Pass 0.50 Pass Pass
16S 1.69 Fail 1.05 Fail Fail
As seen in Table 5.5, both the north and south columns on Pier 13, and the
south column on Pier 16 fail to meet the AASHTO requirements for combined
stresses. Pier 13 fails due to a low allowable axial stress, which results from a long
un-braced length. The south column on Pier 16 fails due to high localized stress in
area of section loss from corrosion.
For continued short-term use, the factor of safety been greatly reduced.
Localized buckling and yielding of the piers presents the greatest danger in the
immediate future. Based on the results of the data collected, Piers 16 and 13 should
be replaced, shored, or strengthened for short-term use, and frequent monitoring
59


should be conducted in the interim period. The areas of corrosion and section loss on
Pier 16 also present additional concerns as likely points of failure through fracture or
plastic deformation.
5.5 Determination of Long-Term Pier Safety
In order to assess the long-term viability of the substructure, a determination
of the effect of maximum temperature variation was made. AASHTO requires steel
bridges in Colorado to accommodate a temperature range of -30 to 120 degrees
Fahrenheit. Using the strain equations generated by linear regression, a bending
stress was calculated for a temperature of -30F for each column. These bending
stresses were then used in lieu of the measured bending stresses in the AASHTO
combined stress equations. The results of this analysis are presented in Table 5.6 and
5.7.
60


Table 5.6 Theoretical bending, calculated axial, and total stresses with
comparison to yield stress for structure temperature of -30F
Pier Theoretical Bending Stress (psi) Axial Stress (psi) Total Stress (psi) % of Yield Strength
3N 17,956 6,324 24,280 67
3S 16,108 6,324 22,432 62
13N 12,862 7,514 20,376 57
13S 11,936 7,514 19,450 54
16N 15,344 14,631 29,975 83
16S 17,711 14,631 32,342 90
Table 5.7 Results of AASHTO Equations 10-42 and 10-43 for combined
axial compression and bending stresses using theoretical bending stresses
Pier Result AASHTO Eq. 10-42 Pass/Fail (Pass if <1.0) Eq. 10-42 Result AASHTO Eq. 10-43 Pass/Fail (Pass if <1.0) Eq. 10-43 Net Result
3N 1.26 Fail 1.27 Fail Fail
3S 1.17 Fail 1.19 Fail Fail
13N 13.33 Fail 1.09 Fail Fail
13S 12.44 Fail 1.05 Fail Fail
16N 1.69 Fail 1.05 Fail Fail
16S 1.88 Fail 1.17 Fail Fail
61


As shown in Table 5.7, many of the columns cannot meet code requirements
for extreme temperatures. Although none of the columns has a total stress larger than
the yield stress, the factor of safety is obviously reduced substantially. This fact is
reflected by the results shown in Table 5.7 where none of the columns in bending
meet the code requirements for combined stresses. These results indicate that the
substructure of the bridge is inadequate for long-term reliability and safety.
62


6. Pier 17 Analysis
6.1 Introduction and Summary
As the only concrete pier in the bridge, Pier 17 required special consideration
during both the strain study and analysis. Pier 17 was the first element of the bridge
to show visible signs of distress. Currently, it is of greatest concern because
reinforced concrete lacks the ductility of the steel and the possibility of catastrophic
failure is more likely. As with most of the bridge piers, Pier 17 was not designed to
handle substantial bending stress. Analysis of Pier 17 based on field data suggests
that the concrete pier is currently performing beyond its intended capacity and
requires replacement.
6.2 Data Analysis
Unlike the steel piers, strain data taken for Pier 17 were converted to crack
widths instead of strains using the equations generated during calibration (see
Appendix C). These data of crack width vs. temperature were plotted as seen in
Figure 6.1, which shows all data points taken during the study. This plot shows a
linear relationship between temperature and crack width. As with the steel piers, 24
and 34 hour monitoring session data were plotted to illustrate the behavior of the
cracks over a short, extended period. These plots can be found in Appendix I.
63


6.3 Determination of Pier Safety
Over the short term, Pier 17 appeared to behave as the steel piers, with a
seemingly linear relationship between crack width and temperature. However, the
extensive flexural cracking of the pier, as well as the width of the cracks suggested
that the reinforcing steel may be overstressed. Localized failures, slippage, and
plastic deformation of the reinforcing bars within the column could not be accounted
for easily with the type and scope of instrumentation used to test the columns.
64


Therefore, a relationship between the measured width of the flexural cracks and steel
stress was calculated using the Gergely-Lutz Equation3:
w = 0.076#, ijdj (6-1)
where w is the crack width, P is the ratio of distances to the neutral axis from the
extreme tension fiber and from the centroid of the main reinforcement (a value of 1.2
is recommended), dc is the depth of clear cover on the primary flexural reinforcing,
and A is the area of effective tension concrete.
The stress in the reinforcing steel, or fs, was calculated for the data on each
column, and two methods were used. For each column, minimum and maximum
crack widths were determined based on field data. The steel stress was then
calculated by Equation 6.1 based on both the measured range of the crack width and
the absolute maximum crack width. These results are presented in Table 6.1.
Table 6.1 Calculated steel stresses based on crack width range and absolute
maximum crack width
Column Minimum Crack Width (in) Maximum Crack Width (in) Crack Range (in) Steel Stress from Range (ksi) Absolute Steel Stress (ksi)
Column 1 0.08340 0.14948 0.06608 122 277
Column 2 0.15799 0.20208 0.04409 24 109
3 Building Code Requirements for Structural Concrete (ACI 318-95), 1995
65


Results presented in Table 6.1 indicate that the reinforcing steel in Pier 17 has
passed the yield point. Since the pier has not yet failed, the validity of using the
Gergely-Lutz equation in this application may be somewhat questionable. Several
factors may be responsible for reducing the accuracy of the equation in this
application:
1. The equation was developed for use with beams and one way slabs, in this
application analysis is performed on a beam-column.
2. Use of the equation has been for guidance in detailing bar layout. It is not
specified for use in back-calculating stresses based on crack widths4.
3. Research indicates that a non-linear relationship exists between the crack
width at the surface of a member, and the crack width at the surface of the
bar. The crack width at the bar has been found to be considerable smaller
that the width at the surface5.
Consequently, the results of the Gergely-Lutz equation in this instance should
serve only as an indicator of the actual stresses present in the Pier. Further modeling
and analysis should be carried out to determine the current factor of safety for the
Pier. Frequent monitoring of the Pier should be performed in the short-term.
4 Wang and Salmon, 1998.
5 Ferguson, 1973.
66


Based on the crack widths measured and the corresponding increased likelihood
of corrosion, the Pier is not viable as a structural member for long-term use and will
have to be replaced or rehabilitated.
67


7. Summary and Conclusion
7.1 Summary of Findings
The study was successful in applying the use of strain gauges to the specific
structural problems of the 6th Avenue Viaduct. Important findings include:
Confirmation of displacement of the bridge superstructure with respect to
temperature changes
Confirmation of bending stresses in bridge piers
Determination of a relationship between temperature and strain in bridge
piers
Current overstress conditions in three columns
Code violations indicating long-term structural safety issues warranting
corrective action
Necessary monitoring over the short-term to insure safety
Current conditions indicate that the bridge does not meet design requirements;
it is safe for short-term use as long as frequent monitoring of the structure is
conducted and some short-term repairs are made. Obviously, the cracking on Pier 17
and the areas of corrosion on Pier 16 present possible locations of catastrophic failure.
These weakened areas should likely be shored or strengthened for short-term bridge
68


use. Long-term viability of the Eastbound 6th Avenue Viaduct will depend on what
large-scale corrective measures are applied. However, it is recommended that a
repair program be implemented soon, as continued temperature cycles will only serve
to exacerbate the current problems with the structure.
7.2 Recommendations for Future Study
Some additions to the monitoring program would have reduced possible errors
in the data, enhanced the findings of the study, and allowed for more accurate
analysis of the bridge. Helpful additions for future similar studies include:
1. Measurement of the structure temperature rather than the ambient air
temperature and development of a methodology to measure an average
temperature across the entire structure. This measurement would more
accurately correlate temperature and strain by reflecting the actual
temperature of the elements important in structural behavior.
2. Use of strain rosettes to determine possible torsional forces or
nonsymmetrical bending.
3. Real-time, continuous strain reading would provide a more complete data
set for analysis. The methodology used in this study provided point data
with significant gaps. Continuous monitoring would have allowed for
69


additional analysis including the effects of rush-hour traffic, as well as
collection of greater amounts of data with less possibility for error.
4. Increased instrumentation of Pier 17 would have provided for enhanced
behavioral data. If all cracks on one column were instrumented, greater
correlation of data and improved analysis would have been possible.
7.3 Future Corrective Repairs
At the time of publication, a structural consultant was preparing a design to
repair the Eastbound 6th Avenue Viaduct for the City and County of Denver. Due to
significant cost the repairs will be phased, dealing with the most severe safety issues
first. The repairs slated to begin in the fall of 2002 will primarily address
substructure deficiencies from Pier 8 west for the eastbound bridge. The repair
includes:
Replacement of straight steel piers 11, 13, 15, 16
Replacement of Pier 17
Addition of helical piers to the foundation of all piers
Replacement of all bearing devices
Strengthening of girder connections at previous hinge locations
The consultant believes that stiffening the substructure of bridge and replacing
the existing bearings will adequately accommodate expansion and contraction.
70


Replacement of Pier 17 includes redesigned round columns with a new footing. Steel
pier columns will be replaced with new, larger sections (strong axis bending turned
along the longitudinal axis of the bridge) and the existing pier cap will be reused.
The footings on all steel piers will be enlarged and strengthened.
71


Appendix A
Typical Views of the 6th Avenue Viaduct
72


Eastbound 6th Avenue Viaduct looking east from west abutment
73


74


View underneath 6th Avenue Viaduct in Union Pacific rail yard
i
75


76


East bridge abutment
77


I
Frozen girder joints with welded steel plates
78


Pier 8 A-Frame configuration
79


Pier 17 looking west toward abutment
80


Corrosion and section loss at the base of Pier 16 North
81


Corrosion and section loss at the base of Pier 16 North
82


West bridge abutment
83


Appendix B
Strain Gauge User Manual
84


Instruction Manual for Installation of Strain Gauges
and Operation of Strain Reading Equipment
Prepared for: Kevin L Rens, PhD, PE
By: Benjamin Allen
May 2001
85


Preface
This manual was prepared for Dr. Kevin L Rens as an aid for use with the P-
3500 Digital Strain Indicator (see Figure 3.2.1) and the SB-10 Switch and Balance
Unit (see Figure 3.3.1). The knowledge contained in this manual has been gained
largely by experience and research of the author. It is the authors hope that the
contents of this manual will enable others to use strain gauges and the accompanying
necessary equipment with greater ease and efficiency.
The procedures described herein for gauge preparation and installation are
those used in one application for one type of gauge. The set-up described produced
satisfactory results for the application. As other uses and conditions will undoubtedly
vary, alternate arrangements will likely be required or preferred. It is therefore
recommended that the procedures and instructions of this manual be viewed as a
useful reference or guide, and it is further recommended that future users review the
referenced literature before any major work is done.
While the instructions for preparing and installing strain gauges are specific to
a single application, the operating instructions for the P-3500 and SB-10 are
universally applicable for a quarter bridge set-up.
86


Table of Contents
Chapter Number Page
1. General Strain Gauge Information 1
2. Strain Gauge Preparation Installation and Protection 6
3. Set-up and operation of Strain Reading Equipment 24
Appendix
A References 34
B Strain Gauge Verification Using a Cantilever Beam 42
87


Chapter 1: Strain Gauge Basics
1.1 Introduction
Strain Gauges are among the oldest non-destructive testing technology, dating
back more than fifty years. They are used to measure changes in strain in many types
of materials including concrete, metals, rock, ceramics, masonry, glass, wood and
plastics. The fact that they are still used today is a testament to their reliability, ease
of use, and versatility. The technology upon which strain gauges are based is
conceptually simple and fairly practical for both laboratory and field use; requiring
only a general understanding of the physics of circuits mechanics, and a few pieces of
equipment.
1.2 Strain Gauge Description
A strain gauge is a small electrical resistor consisting of a thin wrapped wire
that is laminated between two thin plastic sheets (see Figure 1.1). Strain gauges come
in different sizes and types. The material being evaluated and the conditions under
which the evaluation will occur generally dictate the size and type of gauge used.
Some strain gauges, like the ones used by the author, measure strain in one direction
only, while other types of gauges, called rosettes are capable of measuring strain in
several directions.
88