fl s.c.r. t.o....
TRANSCRIPT
REPORT NO. UMTA - FL - 06 - 0017 - 80 -1
S.C.R. T.O. LIBRARY
' STATIC AND DYNAMIC TESTS
OF FULL-SCALE DOUBL_E-TE_E GIRDERS
FOR DADE COUNTY RAPID TRANSIT SYSTEM
METROPOLITAN DADE COUNTY
OFFICE OF
TRANSPORTATION ADMINISTRATION
OCTOBER 10, 1979
This document is available to the U.S. public through the National Technical Information Service,
Springfield, Virginia 22161
PREPARED FOR
RECEIVED
JUL 11 1\jso uaRARY..
U.S. DEPARTMENT OF TRANSPORTATION
Urban Mass Transportation Administration
Office of Technology Development and Deployment
Office of Rail and Construction Technology
Washington, D.C. 20590
NOTICE
This document is disseminated under the sponsorship of the Department of Transportation in the interest of information exchange. The United States Government assumes no liability for its contents or use thereof.
NOTICE
The United States Government does not endorse products or manufacturers. Trade of manufacturers' names appear herein solely because they are considered essential to the object of this report.
STATIC AND DYNAMIC TESTS
OF FULL-SCALE DOUBLE-TEE GIRDERS
FOR DADE COUNTY RAPID TRANSIT SYSTEM
October 10, 1979
The preparation of this report has been financed through a grant from the U. S. Department of Transportation, Urban Mass Transportation Administration.
0 1829
TA 683.9 +H"78
1. Report No . 2 . Governm~nt Accession No .
UMTA-FL-06-0017-80-1 4 . Title ond Subtitle
Static and Dynamic Tests of Full Scale Double-Tee Girders for Dade County Rapid Transit System
Technical Report Documentation Page
3. Recip ient ' • Cotolo11 No .
5. Report Dote
October 10, 1979 6. Performing Or11onizotion Code
t--:e----------------------------------1 8 . Performing Orgon, zotion Report No . 7. A.,thor 1 •l
Thomas T. C. Hsu, Phd. P.E. 9, Performing Or11ani1otion Nome ond Addreu 10. Work Un,t No. (TRAIS )
Met ropo 1 i tan Dade County 11. Controcl or Gron! No .
94f~~eF9f 1~~n~~~rtatt~n9~?~inistration FL-06-0017 ~ __ M_i_a_m_i _,_F_Y_2_r_i_d_a __ -:>_3_T_~_6 _________________ .....J 1 J . T YP• of Report ond Period Covered
12. Sponsoring Agency Name ond Address
U. S. Department of Trqnsportqtion Urban Mass Transportation Administration Final
8ffffice off Technoldogy Develop~ent and Deployment ice o Rail an Lonstruction Technology 14 . Sponsoring A11ency Cod•
400 7th Street, S.W., Washington, D.C. 20590 IS . S.,pplementory Notes
16. Abstract ---------------------------------------i Demonstration tests were carried out on three full-size 80 feet long by 5 feet deep by 12 feet wide precast prestressed double-tee girders proposed for the aerial guideways of Dade County Rapid Transit System; use of this cross-section of girder was a first for any U.S. Transit System. Two of the three girders cast in Miami were shipped by rail to P.C.A. Laboratories, Skokie, Illinois for extensive static and dynamic testing, with the third girder being kept in
Miami to monitor camber and loss of prestress. Static test results of uncracked and deliberately precracked girders, showed that service torsional rotations, which were in close agreement with a theoretical mixed torsion analysis, were small enough to ensure rider comfort. Fatigue resistance of the girders was fully established by two separate test spectra involving 5 and 6 million cycles of loading representing the cumulative damage of sixty years of operational life; no deterioration whatsoever was observed 1n terms of flexural and torsional stiffness, crack propagation or strand stresses. Post-cracking behavior of the girders showed adequate strength and ductility with an ultimate capacity of 1.6 times the required factored severe derailment loading including 100% impact. This test report concludes that the excellent behavior of the girders both from serviceability and strength considerations, substantiated all the design analyses and details used.
17. Key Words
Full Scale Tests, Double-Tee Girders Design Analysis, Double-Tee Girders Static and Dynamic Tests, Precast Girders
18. Distribution Statement
19. $ec.,,ity Clouif. (of thia report) 20 . Security Clouil. (of thi a pogel
Fo,111 DOT F 1700.7 !8-721 Reproduction of completed poge authori&ed
21, No. of Pages 22. Price
151
CONTENT
1. INTRODUCTION
2. BACKGROUND
2.1 VEHICLE LOADINGS
2.2 DESIGN CRITERIA
2.2.1 Live Load
2.2.2 Superimposed Dead Load
2.2.3 Fatigue Load for Test
2.3 DESCRIPTION OF GIRDERS
2.4 FLEXURAL DESIGN
2.5 TORSIONAL DESIGN
2.6 DESIGN OF DETAILS
2.7 PRODUCTION OF GIRDERS
2.8 TRANSPORTATION OF GIRDERS
3. SCOPE OF TESTS
4. MATERIALS
4.1 CONCRETE
4.2 PRESTRESS STRANDS
4.3 MILD STEEL REINFORCEMENT
5. TEST SET-UPS
5.1 GIRDER 2
5.1.1 Initial Static Test
5.1.2 Flexural Fatigue Test
5.1.3 Flexural Derailment Test
5.2 GIRDER 3
5.2.1 Torsional Fatigue Test
5 . 2.2 Torsional Derailment Test
1
3
3
3
3
5
6
7
8
9
10
11
11
12
14
14
16
17
18
18
18
18
19
19
19
20
6.
7.
8.
INSTRUMENTATION
6.1 LOADS
6.2 TRANSLATIONS
6.3 TORSIONAL ROTATIONS
6.4 STRAIN GAGES
6.5 CRACKS
6.6 RECORDINGS
6. 7 LOSS OF PRESTRESS
6.8 CAMBER GROWTH
TEST PROGRAMS AND PROCEDURES
7.1 GIRDER 2
7.1.1 Initial Static Test
7. 1.2 Flexural Fatigue Test
7. 1.3 Flexural Derailment Test
7.2 GIRDER 3
7. 2 .1 Torsional Fatigue Test
7.2.2 Torsional Derailment Test
TEST RESULTS AND ANALYSIS
8.1 GIRDER 1
8.1.1 Loss of Prestress
8.1.2 Camber Growth
8.2 GIRDER 2
8.2.1 Natural Frequency
8.2.2 Initial Static Test
8.2.2.1 Vertical Deflections
8.2.2.2 Torsional Rotations
8.2.3 Flexural Fatigue Tests
8.2.3.1 Stiffness
8.2.3.2 Cracks
8.2.3.3 Strains in Strands
21
21
21
22
22
23
23
23
25
26
26
26
27
29
30
30
31
32
32
32
32
32
32
34
34
35
37
37
37
38
8.2.4 Flexural Derailment Test
8.2.4.1 Vertical Deflections
8.2.4.2 Torsional Rotations
8.2.4.3 Ultimate Load
8.2.4.4 Strains in Strands
8.2.4.5 Crack Development
8.2.4.6 Midspan Diaphragm
8.2.4.7 End Diaphragms
8.2.4.8 Sectional Deformation
8.2.4.9 Construction Details
8.3. GIRDER 3
8.3.1 Torsional Fatigue Tests
8. 3.Ll Stiffness
8. 3. 1. 2 Cracks
8.3.1.3 Strains in Stirrups
8.3.2 Torsional Derailment Test
8.3.2.1 Vertical Deflections
8.3.2.2 Torsional Rotations
8.3.2.3 Ultimate Loads
8.3.2.4 Strains in Stirrups
8.3.2.5 Crack Development
8.3.2.6 Re-entrant earner of Dapped End
8.3.2.7 Joint Between Stem and End Diaphragm
8.3.2.8 End Diaphragm
8.3.2.9 Midspan Diaphragm
8.3.2.10 Sectional Deformation
9. BENDING AND PUNCHING TEST OF FLANGE SLAB
9.1 OBJECTIVES
9.2 SET-UP AND INSTRUMENTATION
9. 3 TEST PROGRAM
9.4 TEST RESULTS
9.5 SUMMARY STATEMENTS
Page
39
39
40
41
42
42
43
44
44
44
45
45
45
45
45
45
45
46
47
47
48
49
49
49
49
50
51
51
52
52
53
54
10. CONCLUSIONS
11. RECOMMENDATION
12. REFERENCES
13. METRIC CONVERSION TABLE
TABLES
1. Locations of Vertical Translation Measurements
2. Initial Static Test Program - Girder 2
3. Target Fatique Test Program - Girder 2 (6-million cycle)
4. Actual Fatique Test Program - Girder 2
5. Derailment Test Program - Girder 2
6. Loading to Crack Girder 3
7. Fatique Test Program - Girder 3
8. Derailment Test Program - Girder 3
9. Results of Flange Slab Test
FIGURES
1. Double Tee Girders Tested (Drawings la. - lh.)
2a. Form for Manufacture of Girders
2b. Manufacture of Girders
3. Transportation of Girders by Railroad
4. Hauling Girders Into Laboratory
5. Stress-Strain Curve of Floriaa-Oolite Conc1ete
6. Secant Modulus of Elasticity as Function of Stress Levels
7. Stress-Strain Curve of Prestress Strands
8. Stress-Strain Curve of Mild Steel Reinforcement
9. General View of Initial Static Test - Girder 2
10 . Hold-Down at End Supports
55
58
59
60
FIGURES
11. Loading Locations - Girder 2
12. Fatique Test Set-up
13. General View of Fatique Test •· Girder 2
14. General View of Derailment Test - Girder 2
15. Loading Location - Girder 3
16. Locations of Translation Measurements
17. Measurement of Deformation of Cross Section
18. Locations of Rotation Measurements
19. SR4 Gage Locations in Girder 3
20. Temperature Change of Concrete at Early Age
21a. Crack Pattern for Initial Static Test of ·- Girder 2
21b. Crack Pattern Prior to Fatique Test - Girder 2
22. Fatique Test Program - Girder 2
22a. Cracking Pattern Prior to Fatique Test - Girder 3
23. Loss of Prestress - Girder 1
24. Comparisonof Experimental and Theoretical Prestress Lossess - Girder 1
25. Camber Growth - Girder 1
26. Loads vs. Midspan Deflections (uncracked) - Girder 2
27. Load vs. Midspan Deflections (cracked) - Girder 2
28. Deflections Along Girder Axis - Girder 2
29. Torque vs. Midspan Rotation (uncracked) - Girder 2
30. Torque vs. Midspan Rotation (cracked) - Girder 2
31. Rotations Along Girder Axis - Girder 2
32a . Load vs. Midspan Deflection (center line) - Girder 2, Derailment Test
32b. Load vs. Midspan Deflection (center line) - Girder 2, Derailment Test
33. Deflection Along Center Line Axis - Girder 2, Derailment Test
FIGURES
34. Load vs. Midspan Deflection (Heavily-Loaded Stem) - Girder 2, Derailment Test
35. Load vs. Midspan Deflection (Heavily - Loaded Stem) - Girder 2, Derailment Test
36. Deflection Along Heavily-Loaded Stem - Girder 2, Derailment Test
37. Torque vs. Midspan Rotation - Girder 2, Derailment Test
38. Torque vs . Midspan Rotation - Girder 2, Derailment Test
39. Torsional Rotation Along Girder 2, Derailment Test
40. Load vs. Strand Strains - Girder 2, Derailment Test
41. Crack Pattern of Girder 2 After Derailment Test
42a. Development of Cracks at Midspan - Girder 2, Derailment Test
42b. Development of Cracks at Midspan - Girder 2, Derailment Test
43. Mechanism Creating Horizontal Cracks at Midspan
44. Diagonal Cracking of End Diaphragm - Girder 2, Derailment Test
Cracking of Stem at End Diaphragm - Girder 2, Derailment Test 45.
46.
47.
48 .
49 .
50.
Load vs. Mids pan Deflection (center 1 i ne) - Girder 3, Derailment Test
Load vs. Midspan Deflection (center 1 i ne) - Girder 3, Derailment Test
Deflection Along Center-Line Axis - Girder 3, Derailment Test
Load vs. Midspan Deflection ( Heavily-Loaded Stem) - Girder 3, Derailment
Load vs. Midspan Deflection (Heavily-Loaded Stem) - Girder 3, Derailment
51. Deflection Along Heavily Loaded Stem - Girder 3, Derailment Test
52. Torque vs. Midspan Rotation - Girder 3, Derailment Test
53 . Torque vs . Midspan Rotation - Girder 3, Derailment Test
54. Torsional Rotation Along Girder 3, Derailment Test
55. Stresses in the Outer Legs of Stirrup in the Heavily-Loaded South Stem -Girder 3, Derailment Test
56. Stresses in the Inner Legs of Stirrups in the Heavily-Loaded South Stem -Girder 3, Derailment Test
Test
Test
57. Stresses in the Outer Legs of Stirrups in the North Stem - Girder 3, Derailment Test
FIGURES
58. Stirrup Stresses in the End Diaphragm - Girder 3, Derailment Test
59. Flexural Cracking Pattern - Girder 3 at 5.0 x Crush Liv~ Load (2.5 x Derailment Load)
60. Shear Cracking Near Midspan - Girder 3, at 5.0 x Crush Live Load (2.5 x Derailment Load)
61. ·Cracking Pattern at the Heavily-Loaded End - Girder 3
62. Crack Pattern at the Joint Between Stem and End Diaphragm - Girder 3 at 5.0 x Crush Liye Load (2.5 x Derailment Load)
63. Crack Pattern of End Diaphragm - Girder 3 at 5.0 x Crush Live Load (2.5 x Derailment Load)
I
64. Crack Pattern of Midspan Diaphragm - Girder 3 at 5.0 x Crush Live Load (2.5 x Derailment Load)
65. Local Horizontal Cracks at Midspan - Girder 3 at 5.0 x Crush Live Load (2.5 x Derailment Load)
66. Location of Wheel Loads in Derailment Condition
67. Instrumentation in Flange Slab Test
68. Load vs. Deflection Curve for Flange Slab Test
69. Crack Pattern After Failure at 1.6 x Crush Live Load
70. Punching Shear Failure at Load Point H
71. Punching Shear Failure at Load Point G
1
INTRODUCTION
Metropolitan Dade County, Florida, is now in the process of building a new rapid transit system of about 50 miles. The first-stage construction of this new system scheduled for completion in 1984, consists of 22.5 miles, and includes 21.5 miles of aerial guideways. The aerial structures to be used for the guideways are prestressed concrete double-tee girders - a first in the use of such a structure in a U. s. transit system.
The double tee design concept resulted from a study conducted by Kaiser Transit Group, the General Architectural and Engineering Consultant for Dade County Rapid Transit System. The study examined various cross-sectional configurations in different types of materials and alternative methods of construction, and narrowed the field down to the traditional box section and double tee section. The double tee girder was recommended for reasons of aesthethics as well as economy. It was estimated that the use of double tee may result in savings of up to $5 million.
It was decided at that stage to conduct more detailed technical studies, examining the various aspects of design and performance of the structure. It is recognized that torsional moments in a transit guideway are produced by wind load on the vehicles and structure, by the transverse horizontal nosing/lurching action of the vehicles on the rail, centrifugal force on curved track, etc •• Torsional rotation caused by these moments must be, therefore, controlled to provide riding comfort for passengers. This is accomplished through the analysis of the torsional stresses and the design of the structure accordingly. Such a study was conducted for this project by Dr. T. c. Hsu, and a report entitled "Torsion Analysis of Dade County Rapid Transit Aerial Guideways" was completed in April 1977. It showed that the torsional rotation can be controlled well within the prescribed tolerance limits.
The elastic torsional analysis, however, is applicable to the behavior of structures before cracking. The post-cracking torsional behavior of doubletee sections was not completely clear at this stage of investigation. However, it was known that large torsional moments occur in the case of severe derailment. For these reasons, testing of two approximately 5/8 scale models was conducted. These model girders were manufactured at Stresscon International, Inc., Miami, Florida and tested at the Construction Technology Laboratories of the Portland Cement Association, Skokie, Illinois in July, 1977. Another study by T. C. Hsu entitled "Torsion Tests of Double-Tee Girders for Dade County Rapid Transit Aerial Guideway" was completed in August, 1977. Test$ conducted under this study showed that the girders behave satisfactorily, both from a safety as well as operational point of view.
In view of large capital investment involved, and the fact that a double tee had not been used on any rapid rail transit system, the U. s. Department of Transportation's Urban Mass Transportation Administration (UMTA) agreed to fund full-scale tests through a Research, Development and Demonstration (Section 6) grant. The full-scale tests were to serve three purposes: -to prove the adequacy of all design methods, to check some of the construction and reinforcement details, and to verify the dynamic performance of the girders.
2
On March 10, 1978 a Structural Design Review Panel of seven eminent engineers was convened to examine the proposed design of the double-tee girders. This panel unanimously endorsed the double-tee design concept and urged the full-scale testing as proposed by the Kaiser Transit Group.
Preparatory work on the testing of full-scale girders began immediately after the recommendation of the Review Panel. Three girders were manufactured by Lone Star Florida, Inc./Stresscon from August to October of 1978. Two of these girders were shipped to the Construction Technology Laboratories of the Portland Cement Association and tested during the period from November 1978 to April 1979. This report describes the testing of these three girders and the analysis of the test results.
Finally, it is to be noted that on August 22, 1979, when the bids for the fabrication of the standard girders for the 21.5 mile phase 1 were opened, all the three bidders opted for the double tee alternative. Furthermore, the low bidder, R. T. Joint Venture, was $1.6 million below the engineers estimate. Since the box girders alternative was not bid on, direct comparison cannot be made. However, it is estimated that the selection of the double tee girders reduced the construction cost of the aerial guideway by about $6.2 million.
2 • BACKGROUND
2.1 VEHICLE LOADINGS
The double tee girders were designed to carry a train of cars,
each 75'-0" long, 10'-6" wide and 12 1 -1 11 high. Each car has two trucks
with a center-to-center distance of 54'-0". Each truck consists of two
axles 6'-6" apart.* Each axle has two wheels suitable for a track guage
of 4 ft. 8-1/4 inc. The empty weight of the car is 78 KIPS, and the
girders were designed to resist a normal full live load of 102.9 KIPS
and a crush live load of 115.S KIPS. The vertical impact factor is
assumed to be 1.25 and the horizontal nosing/lurching force is taken as
10% of the vertical load; centrifugal forces are limited to a maximum of
6.3% of the design axle load (excluding impact) for curved track with a
radius of 5200 feet.
2.2 DESIGN CRITERIA
2.2.1 Live Load
The design criteria used for live load in the tests are:
(1)
(2)
k Normal service load for tangent track (72 vert. load
k-ft + 31 torque/truck): 100% crush live load+ 25% impact
+ 10% nosing/lurching+ 27 mph wind.
K : Abnormal service load for tangent track (72 vert~ load
+ 63k-ft torque/truck): 100% crush live load+ 25% impact
+ 10% nosing/lurching+ 60 mph wind.
(3) Abnormal service load for curved track with radius of 5,200
k k-ft ft. (72 vert. load+ 90 torque/truck): Abnormal service
load (2) + eccentricity of chorded construction for 5200 ft.
radius+ centrifugal force of vehicle with 70 mph speed.
*The distance between axles of 6 1 -6 11 was later changed to 7'-6" which was used in the flange slab test of Chapter 9.
3
(4) Normal fatigue load on tangent track (64k vert. load+
22.5k-ft torque/truck): 89.1% crush live load+ 25%
impact+ 10% nosing/lurching+ 13.5 mph wind.
(5) Abnormal fatigue load on tangent track (72k vert. load_+
25k-ft torque/truck): 100% crush live load+ 25% impact
+ 10% nosing/lurching+ 13.5 mph wind.
(6) Abnormal fatigue load for curved track with radius of
5,200 ft. (72k vert. load+ 48.5k-ft torque/truck):
Abnormal fatigue load (5) + Eccentricity of chorded con
struction for 5,200 ft. radius+ centrifugal force of
vehicle with 70 mph speed.
(7) Severe derailment load on tangent track (115.5k vert.
load+ 346k-ft torque/truck): Crush live load with 100%
impact and maximum side shift of 3 ft.
(8) Factored derailment load on tangent track (162k vert. load
+ 485k-ft torque/truck): 1.4 times severe derailment load
(7).
These criteria were checked in the following tests of Girders 2 and 3:
GIRDER
2
3
TESTS
Initial Static Test
Flexural Fatigue
Flexural Derailment
Torsional Fatigue
Torsional Derailment
CRITERIA USED
(1), (2), (3)
(4), (5)
(7) , ( 8)
(6)
(7), (8)
Flange Bending & Punching (7), (8) (Derailment)
4
2.2.2 Superimposed Dead Load
In addition to the live load caused by vehtcles and wind, the
weight of the track rails, rail plinth pads, power rail, guard rail,
cableway, acoustic barrier, etc., will create superimposed dead
loads (S.D.L.), which is assumed to be 0.81 kip/ft. of tangent track. In
order to simplify the top surface of the test girder, it was
necessary to add 0.04 kip/ft. of concrete. Hence,t~e S.D.L. for the
test beam was taken as 0.77 kip/ft.
The superimposed dead load also creates a uniformly distributed
torque due to the eccentric locations of the acoustic barriers, the
third rail and the cableway. This torque is assumed to be 0.50 kips
ft/ft. of track.
In the case of derailment, it is assumed that the superimposed
dead load is 1.1 (S.D.L.) + 0.1 (Girder weight of 2.31 kip/ft.). It
is also assumed that the acoustic barrier (0.185 kip/ft. weight and
6.17 ft. from shear center) has been pushed off the girder during de
railment. Consequently, the uniform vertical load becomes 0.88 kip/ft.
and the uniformly distributed torque is 0.71 kip-ft/ft. This torque
is in the direction opposite to the derailment torque.
In the testing of Girders 2 and 3, S.D.L. is simulated by two
pairs of concentrated loads. In the flexural test of Girder 2, these
concentrated loads are determined such that they create at the SR-4
gage locations (3.5 ft. from midspan) a moment identical to that
caused by the uniformly distributed superimposed loads and torques.
In the torsional test of Girder 3, the concentrated loads were de
termined to match the shear at 6 ft. from the support.
5
2.2.3 Fatigue Load for Test
The girders are designed for a service life of 60 years and a
fatigue life of over 16 million cycles of loading during this period.
The number of cycles are distributed as follows:
(1) Abnormal Fatigue Cycle (crush live load+ impact,
P = 144 kips): 250,000 cycles.
(2) Normal Fatigue Cycle (Normal full live load+ impact,
P = 128 kips): 6 million cycles.
(3) Reduced Normal Fatigue Cycle (87.1% normal full live
load+ impact, P = 112 kips): 10 million cycles.
For a 6-car train each of these three fatigue cycles consist of 40%
high cycles (moment range from zero to 58P/4) and 60% of low cycles
(moment range from 25P/4 to 58P/4).
Since the time required to apply 16.25 million cycles of loading
with six types of moment range is prohibitive, it is necessary to de
vise some simple equivalent tests with reduced number of cycles and
one or two types of high cycle moment range. Based on Minor's
hypothesis of cummulative damages, Dr. John Fisher(12) has provided
three equivalent tests of 7, 6 and 5 millions high cycles as f0llows:
Equivalent
Test Spectrum
Abnormal
Fatigue Cycle
Normal Total
Fatigue Cycle Cycles
6
A
B
C
250,000 6. 82 million 7 .07 million
3.20 million 2. 80 million 6.00 million
5.05 million 0 5.05 million
Test Spectra Band C have been selected for Girders 2 and 3, respectively.
2.3 DESCRIPTION OF GIRDERS
The cross section and the details of the test girders is shown in
Figs. la and lb. The girder is 80 ft. long, 12 ft. wide, 5 ft. deep and
weighs 94 tons. It has a 9 in. diaphragm at midspan and diaphragms at
the ends. The <lapped end is 2.5 ft. deep and 1'-5' long and the span
between the centers of bearing is 78.5 ft. The two stems, each having
an average width of 11.5 in., are 5 ft. c. to c •• The flange is 8 in.
thick between the stems and tapered from 8 in. at the stem to 6-1/4 in.
at the edge.
The girder is prestressed with 54 one-half in. diameter Lo-Lax
strands. The strands are held down at two points 10.5 ft. from
midspan. The eccentricities of the centroid of all strands are 35 in.
and 15 in. at the midspan and at the ends, respectively.
The three girders were initially prestressed to different levels
as follows:
Girder
Prestress per Strand (kips)
1
29.2
2
27.6
3
28.5
The prestresses in Girders 2 and 3 were reduced below the level of
Girder 1 so that the prestress levels at the time of fatigue testing
would be close to the effective prestress of Girder 1 after all losses
have taken place. The final prestress loss for Girder 1 is calculated
to be 23%.
The stems are reinforced with No. 4 stirrups at 10 in. spacing.
This spacing is increased to 20 in. between the hold-down points and
7
8
is decreased to 4 in. near the ends. The longitudinal steel consists of 6
No. 4 bars per stem reduced to 2 No. 4 bars per stem in the central 12 ft.
between the hold-downs. In the 3 ft. 5-1/4 in. length from the end, special
reinforcement for the <lapped end of the girder is provided according to the
PCI Design Handbooks.(3)
The transverse reinforcement in the flange consists of No. 5 bars at
8 in. spacing at the top and No. 4 bars at the bottom. These bottom No. 4 bars
are spaced at 8 in. between the stems and 16in. in the overhanging flanges.
This arrangement is accomplished by staggering the bottom bars, each bar extending
to the edge of the flange on one side and terminating just beyond the stem
on the other side. The end of each bottom bar at the edge of the flange is
bent up to form a closed loop with the top bar. The longitudinal bars in the
flange consist of 8 No. 4 bars and 13 No. 3 bars distribt¢ed arotmd the loop.
2.4 FLEXURAL DESIGN
Flexural design is based on the well=known principles of elastic behavior
of prestressed concrete. The allowable stress at the bottom fiber is limited
to 2.25 .{if. under 100% crush live load plus impact, and to zero under normal
full load (89.1% crush live ~~ plus impact. The ultimate flexural strength
of the test girder is 9,700 k-ft. whi~ exceeds the required flexural strength
of 7,620 k-ft. by 27%. In other words, flexural design of this double-tee
girder is governed by allowable stresses.
A detailed analysis of a very similar girder is given in Ref. 4. This
analysis is quite lengthy and includes a sophisticated method for calculation
of pres tress losses suggested by Dhondy. All these tedious design procedures
have been simplified by the development of an e lectronic comput e r program. A
det a iled descripti on of the computer program h as been documented in a report
by the writer.CS) The output of this program includes the transformed section
properties, the maxi mum design momen t , the f l exural stresses, the prestress
losses, the ultimat e moment capacity, the release strength of concrete, the
camber and the de f l e ction due to dea d and live loads.
2.5 TORSIONAL ANALYSIS
An elastic torsional analysis of the girder is given in Ref. 1. This
analysis is based on Vlasov's mixed torsion theory for St.-Venant torsion
and warping torsion.<6) This tedious torsional analysis has also been com~
(1 5) puterized. ' The computer program can generate the following infor-
mation:
(1) Sectional constants in mixed torsion analysis:
This includes the location of the shear center, the
St.-Venant torsional constant, the sectorial moment of
inertia and the dimensionless constant k = J,,/GC/EI w
(2) The deformation and internal forces due to torsion:
Th~s includes the rotation, the rotation per unit length,
the bimoment> the St.-Venant torque and the warping torque.
The end conditions may be fixed (100% warping restraint),
hinged (zero warping restraint) or partially-hinged (actual
dimensions of diaphragms).
(3) Stress diagrams due to unit force:
This consists of the shear stresses due to unit St.-Venant
torque, the shear stresses due to unit warping torque and
the longitudinal stresses due to unit bimoment.
This elastic torsional analysis is applied to design in two ways.
First, to predict the torsional rotation of the girder under service loads,
in order to ensure rider comfort. Second, to design the torsional steel
under service loads.
Vlasov's elastic mixed torsion analysis was extended to the post
cracking stages for the study of derailment problems. This so-called
"post-cracking mixed torsion analysis 11 has been developed by Hwang and
the writer. This analysis has been used to design the torsional steel
9
under severe derailment loads.
The design of stirrups and longitudinal steel caused by shear and
torque was based on the paper by Zia and McGee.< 7) However, the minimum
stirrup requirement was calculated according to a more recent paper by
Zia and the writer.<8)
2.6 DESIGN OF DETAILS
One of the purpose of these tests is to check the suitability of
the construction details. Design of these details can be found in
Ref. 9. Three of these details are of prime concern:
(1) The re-entrcnt corner of the clapped ends:
Two bearing plates, each 3/4 in. by 6-3/4 in. by 16 in.,
are used at each end of the girder. Each bearing plate
was anchored into the concrete by 4 No. 5 vertical bars and
8 horizontal bars (2 No. 9, 2 No. 7, 2 No. 6 and 2 No. 4).
At the suggestion of Professor Ben Gerwick, this plate was
extended into the end diaphragm by 3/4 in. to meet the
closed galvanized stirrups of the end diaphragm. In this
way the diagonal cracks usually developed at the reentrant
corner may be arrested.
(2) The hold-down points:
Each hold-down device is designed to resist a hold-down
force of 47 kips. This force introduces large stress con
centration in the strands and in the concrete.
(3) The connections between the stem and the end diaphragm:
These regions require particular attention because (a)
warping torsion creates tension in this region and (b) pre
stress is not fully effective in regions so close to the end.
10
To reinforce this region, two additional bars, (1 No. 8 and
1 No. 6, each 8 ft. long), are placed at the bottom of each
stem near the end of the girder. Another fours No. 6 horizon
tal bars are placed at the bottom of each end diaphragm. Two
of the four are bent and anchored into the stems. Similarly,
two No. 6 U-shape horizontal bars connecting an end diaphragm
to the stems are placed at approximately the mid-height of
the girder.
2.7 PRODUCTION OF GIRDERS
The girders were manufactured at Lone Star Florida, Inc./Stresscon.
A self-stressing reinforced concrete form with steel lining was designed
by Givens as shown in Fig. 2a. The process of manufacturing the girders
is shown in Fig. 2b. Details of the production process are given in
Ref. 9.
2.8 TRANSPORTATION OF GIRDERS
The girders were transported by railroad from Miami, Florida, to
Skokie, Illinois (Fig. 3). They were then trucked from the Skokie
siding to the Construction Technology Laboratories of the Portland Cement
Association. The process of threading a girder into the laboratory is
shown in Fig. 4.
11
3. SCOPE OF TESTS
The scope of tests is swmnarized in the following Table:
GIRDER
1
2
3
LOAD CONDITION
None
None
Coupler at Midspan
Coupler at O. 21£
Coupler at 0.25j
OBJECTIVE OF TEST
a) Loss of Prestress
h) Camber Growth
a) Natural Frequency
b) Flexural and Torsional Stiffnesses
c) Flexural Fatigue
d) Flexural Derailment
a) Torsional Fatigue
b) Torsional Derailment
12
Load at edge of flange c) Flange Bending & Punching
A total of three girders were manufactured. Girder 1 was set-up in the
yard of Lone Star, Florida, Inc./Stresscon to monitor the loss of prestress
and the camber growth. Measurements were made periodically, about daily
in the beginning and then gradually increased to monthly in the later stage.
Girders 2 and 3 were shipped to the Construction technology Labora
tory of Portland Cement Association, Skokie, Illinois. A simple test
was first made on Girder 2 to measure the natural frequency of the girder
13
in flexure and in torsion. Girder 2 was then subjected
to loadings (symmetrical about midspan and eccentric with respect to center-
line girder) to maximize the bending moment and the torsional rotation at
midspan. The initial static testing was performed to study the flexu-
ral and torsional stiffnesses under design criteria (1), (2) and (3).
It was first carried out on the uncracked girder, and then on the
cracked girder after it was subjected to a vertical load exceeding twice
the normal service load. Following the static test, the girder was sub-
jected to 6-million cycles of fatigue load according to design criteria
(4) and (5). Finally the girder was loaded under derailment condition
to check design criteria (7) and (8). This loading was increased to
ultimate capacity.
Girder 3 was subjected to loads on one-half span only. The load
closest to the support was placed at a distance of 6.25 ft. therefrom.
These load locations were chosen to satisfy two investigations simultane
ously. First, to allow diagonal cracks, if any, to develop in the stem
near the support under maximum shear and torsion. Second, to maximize
as much as possible the shear at the re-entrant corners of the clapped
ends. Girder 3 was first loaded statically in flexure to cracking before
subjecting it to 5-million cycles of fatigue loads according to design
criterion (6). It was then loaded to design criteria (7) and (8) under
derailment conditions. This loading was increased to near ultimate
capacity until the yielding of the stirrups.
An additional test was also made on Girder 3 to study the bending
and punching capacities of the overhanging flange. This test is reported
independently in Chapter 9.
4. MATERIALS
4.1 CONCRETE
The composition of the concrete mix was:
Cement: ASTM Cl50 Type III
Coarse Aggregate: ASTM C33 (Miami Oolite)
Sieve Opening
% passing by wgt.
l"
100
3/4"
99
Pea Rock: ASTM C33 ~iami Oolite)
Sieve Opening
% passing by wgt.
1/2"
100
3/ 8"
97
1/2"
64
113
37
Fine Aggregate: ASTM C33 (Miami Oolite Screening)
Fineness modulus = 2.6.
Retardant: Daratard HC. ASTM C494. Type D Air-Entrainment Agent: None. Design mix per cubic yard*
Girder 1
Cement 705 lbs.
Coarse Aggregate 1600 lbs.
Pea Rock None
Fine Aggregate 1280 lbs.
Water 36 gals.
Retardant 160-175 oz.
Slump 3-14 - 3-1/2
3/8"
32.6
tt4
12
114
3.1
1110
3.5
Girders 2 and 3
680 lbs.
800 lbs.
800 lbs.
1280 lbs.
33 gals.
160-175 oz.
3" - 4"
*Concrete mix for Girder 1 was found to have workability less than
desirable. The mix was then improved for Girders 2 and 3.
14
118
1.8
1116
2.5
The cylinder compressive strength of concrete is summarized in the
following Table: •
Girder Approx. 1 day Approx. 28 days At Test
15
1
2
3
4209 (16.5-19.5 hr.)
4341 (19-20.8 hr.)
4097 (19.3-21 hr.)
7572 (34 days)
7597 (28 days)
7598 (28 days)
7950 (53 days)
7953 (124 days)
Note that each value, in psi, in the Table is an average of 5 or 6 cylin
ders, each taken from a truck load. The values in brackets are the actual
age of concrete tested.
The stress-strain curve of the concrete at test for Girder 2 is plotted
in Fig. 5. This curve is an average for five cylinders. The secant modulus
of elasticity, E, of this concrete is plotted in Fig. 6 as a function of C
the level of stress. It can be seen that the secant moduli are 3850 ksi and
3410 ksi at 0.1 f' and 0.5 f', respectively. When compared to the modulus C C
of elasticity recommended by the ACI Building Code 318-77,
(E = 57 000 'f' = 5080 ksi), the ratios E /E d are 0.76 and 0.67 for code ' ✓ ~c c co e
stress levels of 0.1 f~ and 0.5 f~, respectively. This low modulus of
elasticity of local concrete is caused by the low modulus of elasticity of
Florida Oolite aggregates. (lO)
The direct tensile strengths of concrete were also measured. A 6 in.
by 12 in. standard concrete cylinder was cut at both ends and was cleaned
carefully. Two thick steel plates were epoxied onto the ends of the
cylinder in a specially designed rig. Each steel plate was then attached
to a spherical joint assembly and was tested in a universal testing
machine. The eleborate processes of preparing the concrete specimens
and the use of the spherical joints were to ensure concentric tensile
loading of the conc~ete cylinder and to avoid failure at the interfaces
between the concrete and the steel plates. Even with such precuati•ons,
however, some cylinders still failed at the interfaces. Since the test
results of such cylinders were conservative, these were included in the
average if they have higher strength than those cylinders with failure in
the concrete. These were neglected, however, if they have lower strength.
The av~rage direct ten•sile strengths of concrete of three or four cylinders
were:
~irder 2: 396 psi at 68 days
Girder 3: 444 psi at 141 days
4.2 PRESTRESS STRANDS
1/2 in. Lo-Lax prestressed strands, Grade 270 and conforming to
ASTM A416, were used in the girders. Six pieces of strands randomly
selected were tested in a 200,000 lb. Wiedemann-Baldwin universal
testing machine at the University of Miami. Steel discs were care
fully glued onto one wire of a strand and the strains were measured by
a mechanical Demec gauge over a length of 8 in. A typical stress
strain curve of strands is given in Fig. 7. The average modulus of
elasticity was found to be 28,900 ksi and the average ultimate
strength was 278 ksi. It is interesting to note that all the wires of
the six strands broke at the jaw. The maximum strain was only 2% - 3%.
It appears that Lo-Lax strands have somewhat less ductility than the
stress-relieved strands, which usually do not break at the jaw.
16
Strains in the prestress strands were also measured by epoxying
SR4 electrical strain gages onto one wire of each ,strand. It was found
that the average modulus of elasticity was 33,700 ksi, considerably
higher than those obtained by Demec gage. Since identical method was
used to measure the strains in the strands of Girder 2, this higher
modulus of elasticity was used to convert the measured strains into
stresses for the strands in this girder.
4.3 MILD STEEL REINFORCEMENT
Mild steel reinforcement, Grade 60 and conforming to ASTM A615, was
used for all non-prestressed reinforcement. 4 pieces of No. 4 bars
randomly selected were also tested. A composite stress-strain curve of
all four specimens is given in Fig. 8. This steel has a long yield
plateau and an even longer strain hardening region. The average modulus
of elasticity is 26,700 ksi and the yield stress is 61 ksi.
17
5. TEST SET-UPS
Both Girders 2 and 3 were simply supported in a similar manner as
shown in Fig. 9. End diaphragms were restrained from twisting as shown
schematically in Fig. 10. The loading systems, however, were different
for the two girders. They are described as follows:
5.1 GIRDER 2
5.1.1 Initial Static Test
Three loading systems have been used to tes.t Girder 2. The
first was for initial static test to study the flexural and torsional
stiffnesses. In this test two pairs of symmetrical loads, each
18
identified as P1 and P2 in Fig. 11, were applied. For flexural moments
without any torsion P1
and P2 should be applied equally. If P1 was increased
while P2
was decreased, a torsional moment will also be generated. The
loads were applied through a system of cross heads and tied rods(ll)
in Fig. 11. A general view of this test set-up has been given in
Figure 9.
5.1.2 Flexural Fatigue Test
The second loading system was used for flexural fatigue test.
Two MTS 110-kip capacity actuators were mounted on the girder as
shown in Fig. 12. Actuators were provided with swivel heads at both
ends. For maximum efficiency of the hydraulic system the piston of
each actuator was clamped to the girder by two cross heads and two
rods prestressed to 60 kips; this prestress will not affect the test
results. Similar to the initial static test, the loads in the girders
were identified as P1 and P2 in Fig. 11. The relative magnitude of loads
P1 and P2 was obtained by controlling the eccentricity of the aC'tuator
relative to the center line of the girder. A general view of this test
set-up is shown in Fig. 13.
5.1.3 Flexural Derailment Test
The third loading system was designed for derailment test. Four
loads, designated as P3
in Fig. 11, represent the center-lines of
19
four axle loads of two derailed cars with the coupler located at mid-span;
the other axles of the two cars would be located off the span. These
loads are located 3 ft. from the center line of the girder. To accommodate
the large deflection required to test the girder to failure, a system
of croDs heads, high strength Dywidag tie rods and high capacity rams
were used. This system could take up the stroke of the rams after •
they are fully extended, so that the successive strokes of the rams
could be added. In addition to the large derailment loads P3
, four
small loads P1
and P2
were also applied to simulate the superimposed
dead load on the girder. These loads were also applied through cross
heads and rods and were maintained constant throughout the test.
A general view of the test set-up is shown in Fig. 14.
5.2 GIRDER 3
5.2.1 Torsional Fatigue Test
Two loading systems have been used to test Girder 3. The first
was torsional fatigue test as shown in Fig. 15b. Two pairs of loads,
each designated by P1 and P2 , were placed 7 ft. and 28 ft. from the end.
These loads were induced by two 110-kip actuators attached to the girder
in the same way as Girder 2, (Fig. 12).
The girder was cracked before the dynamic loads were applied.
Since the two actuators did not have sufficient capacity to crack the
girder, two auxiliary loads identified as P5 were added. P5
was
accomplished by cross-heads, tied rods and a ram. This auxiliary
equipment was removed before the commencement of the fatigue test.
The precracking process produced flexural cracks in the vicinity
of the load P5 in a region 12 ft. wide from 7 ft. west to 5 ft. east
of P5 . A small crack about 5 in. long and inclined at about ·25°
to horizontal was also observed at the south side of the west re-
entrant corner .
5.2.2 . Torsional Derailment Test
The seconG l oading system of Girder 3 was for derailment test
as shown in Fig. 15c. Thi~ load system was identical to that for
Girder 2 except that the locations of the loads were moved closer to
the support.
20
6. INSTRUMENTATIONS
6.1 LOADS
Universal load cells mol.lllted on the MI'S rams were calibrated before
testing and were used to measure loads during the fatigue load tests.
Forces applied at each loading point during the initial static test and
the derailment test were measured by precalibrated compression load cells.
6.2 TRANSLATIONS
Vertical translations relative to the laboratory floor were measured
at 19 points in Girder 2, and at 13 points in Girder 3. The cross sections
where the measurements were made are indicated by alphabet letters in
Fig. lla (for Girder 2) and in Fig. 15a (for Girder 3). Locations of
measurement at each cross section are given by numbers 1 to 5 in Fig.
16a. Using this code the locations of measurement for vertical trans
lations have been sunnnarized in Table 1. Longitudinal translations were
also measured at the ends of the girders as shown in Fig. 16b.
Both vertical and longitudinal translations were measured by potentio
meters with a sensitivity of 0.001 in. Each potentiometer was connected
through hinges to a thin steel plate glued onto the girder. These
hinges eliminated the interference of small movement of the girder in
the plane perpendicular to the direction measured.
Deformation of the cross section during static and derailment tests
were measured at two sections in Girder 2, and at one section in Girder 3.
At each: section a rigid frame was attached to the girder right on top of
the two stems. Eight LVDT or DCDT gages were attached to the frame as
shown in Fig. 17. In Girder 2, one frame was located at quarter point
21
of the girder and the other frame was 10 ft. from midspan in the same
half of the girder. In Girder 3, only one frame was used at the quarter
point in the half span that was loaded.
6.3 TORSIONAL ROTATIONS
Six rotation gages designed by the writer were used to measure the
torsional rotations along the girder. Rotations for Girders 2 and 3
were measured at the locations shown in Fig. 18.
Each rotation gage consists of a heavy 9-lb steel cylinder hung by
a thin strip of spring steel to an aluminum angle, which is attached
to the bottom face of the girder. Four SR-4 electrical strain gages
are epoxied onto both sides of the spring steel strip to form an electrical
Wheatstone bridge. When the girder rotates, the heavy steel cylinder
tends to remain vertical, resulting in bending of the spring steel strip.
The bending strains of the spring steel strip are monitored by the SR-4
gages and the output voltage of thcWheatstonebridge is fed into a re
corder. Precalibration between the output voltage and the torsional
rotation shows that the sensitivity of the rotation gage is 0.01 degree
per division.
6.4 STRAIN GAGES
SR4 electrical strain gages were epoxied to the strands and rein
forcing bars before casting. 24 gages were attached to the strands in
22
both stems of Girder 2 at two symmetrical sections 3.5 ft. from midspan. At
each of the four locations, two gages each were glued to the first and second
strands of the innermost row counted from the bottom. One gage was glued to the
9th strand at the top and the last gage was placed on ei ther the fourth or
sixth strand . To force the cracks to occur at the four gage locations
a crack former was placed at each location between the bottom fibre of
the stem and the bottom strand. All gages were monitored during the
derailment test, but only six gages were selected for recording during
the dynamic test.
31 SR4 gages were epoxied onto the stirrups in the stems and in
an end diaphragm of Girder 3. The locations of these gages are shown
in Fig. 19. Two gages, located nearest to the end of the heavily
loaded stem on the outside leg of stirrups, were selected for periodi
cal reading during the dynamic test. All gages were monitored during
the derailment test.
In the test some gages were fotmd to be dead and the accuracy of
some live gages was also questionable. All tmreliable gages have been
discarded in this report.
6.5 CRACKS
Cracks were marked with black felt pens and photographs were taken.
At some load stages maximum crack widths were selected and were measured
by illuminated hand microscopes.
23
For the dynamic test of Girder 2, one of the largest cracks were
selected for monitoring. An LVDT was attached to two brackets glued onto
the concrete surface on each side of the crack. In this way this crack
width was monitored continuously.
6. 6 · RECORDINGS
Data from load cells, rotation gages, potentiometers and SR4 gages
were recorded by a digital data acquisition system. All the data were
later reduced to engineering units and tabulated by a mini- computer.
6.7 LOSS OF PRESTRESS
Losses of prestress were measured in all three girders.
Losses due to friction, anchorage seating and relaxation of strands
was first measured before casting of concrete. The elastic
24
shortening was then obtained from measurements before and after the cutting
of strands. Loss of prestress due to creep, shrinkage and relaxation was
measured periodically thereafter. The measurements were taken daily in
the early age and the interval was gradually increased to about once a
month at the later age.
Five to ten hours after the concrete was cast in the form, 3/8 in.
di.a.meter steel discs were countersunk and epoxied onto the inside faces
of the two stems at two symmetrical sections 3 ft. - 6 in. from midspan.
To accommodate the installation of those steel discs, four special
windows were provided in the skin of the steel form. These windows
are accessible from the outside by two tunnels under the bed.
Loss of prestress was measured by a Demec mechanical gage and a
Whittemore mechanical gage. At each of the four locations of measure
ment, two steel discs were placed 8 in. apart to accommodate the Demec
gage length and another two steel discs were placed 10 in. apart for the
Whittemore gage. These steel discs were located approximately 5 in.
above the bottom face of the stem at the centroidal line of the prestress
strands. The loss of prestress before and shortly after the cutting of
strands was calculated from the measurements at these four locations.
After a girder was lifted out of the form, steel discs were attached
to four more locations on the outside faces of the two stems, opposite to
those installed previously on the inside faces. The loss of prestress
thereafter were calculated from the measurements at all eight locations.
In view of the fact that temperature fluctuated greatly during the
life of a concrete girder, particularly in the early age as shown in
Fig. 20, the strain caused by temperature must be subtracted from the
strain measured directly by the Demec or Whittemore mechanical gages.
This temperature strain was measured by the same mechanic~l gages from an
auxiliary steel bar with two fixed discs. This st,eel bar was attached to
the face of the concrete at the location of measurement until the tempera
ture of the bar was identical to that of the concrete. To convert the
temperature strain of the auxiliary steel bar to the temperature strain
of concrete we multiply the former by the ratio of the coefficient of
expansion of concrete to that of steel.
6.8 CAMBER GROWTH
Camber was measured innnediately after the girder was removed from
the form and periodically thereafter. Measurements were taken daily in
the early age and the interval has been gradually lengthened to approx
imately once a month.
Measurement of camber was made simply by a long piano wire and a
scale. The piano wire was stretched tightly from both ends of the girder
by hooks and weights. Measurements were made at four locations, namely
under the two stems and under the tips of the two overhanging flanges.
25
7. TEST PROGRAMS AND PROCEDURES
7.1 GIRDER 2
7.1.1 Initial Static Test
The initial static test program for Girder 2 to study the flexural
and torsional stiffnesses of the girder was summarized in Table 2.
The test procedures could be divided into three parts:
(1) Static test of Uncracked Girder (load stages 1 to 26)
(2) Cracking of the girder (Looad stages 27 to 51)
(3) Static Test of Cracked Girder (Load stages 52 to 76)
In the uncracked static test, the 26 load stages can be divided into
six groups:
(a) Apply superimposed dead load (load stages 1 to 5)
(b) Apply bending moments with increment of 6 kips for P1
and P2 (load stages 6 to 11)
(c) Apply torsional moments with increment of 15 k-ft. i.e.,
3 kips force times 5 ft. level arm. (load stages 12 to 17).
(d) Unload torsional moments with 4~crement of 30 k-ft. (load
stages 18 to 20).
(e) Unload bending moments with ·aecrement of 12 kips for P1
and P2 (load stages 21 to 23).
(f) Unload superimposed dead load (load stages 24 to 26).
After the uncracked static test, Girder 2 was cracked. This
process was made with small load increments because crack length was
very sensitive to loads shortly after cracking. Initial cracks were
found at the crack fonners (where the SR4 strain gages were located)
at load stage 29. The actual P1 and P2 were 47.5 kips each and the
26
crack width were less than 0.001 in •. Normal cracking in the constant
moment region between the hold-down points occurred at load stage 35,
with P1 = P2 = 55.7 kips. These loads correspond to a modulus of
rupture of about 350 psi. The maximum crack width was 0.002 in ••
The process to crack the girder was terminated at load stage 50 when
the crack length reached 1.5 ft., P1 = 67.5 kips, P2 = 66.0 kips and
the maximum crack width was 0.006 in .. A composite photo of the
crack pattern at this load stage is shown in Figs. 21a and 21b. This
crack pattern was chosen to satisfy three conditions: (a) The crack
should extend above all the strands in the constant moment region in
order to stabilize the strain measurements of the strands in the
fatigue test. (b) The cracks should not be too long, because the
high sensitivity of cracks at this load stage was desirable to study
the crack propagation in the fatigue test. (c) Cracks should have
started outside the constant moment region. This critical region
just outside the load points P1 and P2 (see Fig. 11) are subjected
to large shear and moment and therefore should be carefully observed.
The third part of the initial static tests from load stages 52
to 76 was identical to the first part from load stages 1 to 26.
These later load stages produced flexural and torsional stiffness
for cracked section.
7.1.2 Flexural Fatigue Test
27
Girder 2 was subjected to 6.02 million cycles of dynamic load according
to Fisher's equivalent Test Spectrum B· at a frequence of approximately
2.5 cycles/sec. The target fatigue test program was given in Table 3.
However, the actual fatigue program was somewhat different as recorded
in Table 4. The differences were due to the change in midtest from
a 7-million cycle program to the 6-million cycle. It was also caused
by the equipment instability of the MTS controller. A comparison of
the target and actual test program was given in Fig. 22. The two
programs should be equivalent according to Minor's Hypothesis of
cumulative damages.
The 6-million cycle test program as suggested by Fisher et
a1< 12) consisted of 3.2-million cycles of 100% crush live load plus
impact according to design criteria (5) and 2.8-million of normal
full load (89.li. of crush live load) plus impact according to design
criteria (4). The torsional moments due to 10% nosing and 13.5 mph
wind were also imposed on the girder. It was assumed that half of
all torsional moments were positive and half were negative. The
combination of two levels of vertical loads and the two directions
of torsional moments resulted in four loading conditions I, II, III
and IV, listed in Tables 3 and 4.
Table 3 showed that both load conditions I and II were split
into two equal parts, one at the beginning and one at the end of the
test. In other words, the heavily-loaded stem will receive the
maximum Pi load of 53.9 kips both at the beginning and at the end of
the fatigue test. Such test history should be conservative if the
load history has an effect on the fatigue strength as some in
vestigators have suggested.
The load history was designed to minimize the number of changes
of the load eccentricity, while maintaining some dispersion of the
four load conditions. This test program required only three changes
of the load eccentricity. (See Fig. 12)
It should also be noted that the superimposed dead loads should
be simulated by P1 = 12.9 kips and P2 = 7.6 kips. However, these were
28
29
changed to P 1 = 12.0 kips and P2 = 8.5 kips in load condition I to
maintain a constant load eccentricity of 5 in •• This increases slightly the
stress rapge in the heavily-loaded stem and, therefore, should be conserva
tive. Similar justifications should apply to the other three load conditions.
In the fatigue test, dynamic correction must be made to account
for the inertia of the girder. Dynamic loads were controlled such that
the deflection produced due to cyclic loading corresponded with the
measured minimum and maximum static deflections. The dynamic deflections
were measured by LVDT's (Linear Variable Differential Transformer) at
midspan. Repetitive load varied sinusoidally from minimum to maximum.
Loads and deflections were monitored periodically throughout the test;
also, the static tests were repeated to establish the effect of creep
deflection.
7.1.3 Flexural Derailment Test
The derailment test program for Girder 2 is summarized in Table 5.
Loads P1
and P2 were calculated such that these loads, combined with the
self weight of the girder, simulated the required factored self and
superimposed dead loads. Loads P 1 and P2 were first applied and main
tained constant throughout the test. The factored derailment load (see
Fig. 11) was then applied by varying P3
in the following sequences:
(a) P3 was loaded in four increments up to 57.8 kips, (load
stage 9, 10) corresponding to the unfactored, derailment
load (=2 times the crush live load) and the design
criterion (7). It was then unloaded to find the residual
cracking deflections and rotations as a measure of the
serviceability of the girder after a severe derailment.
(b) Repeat (a) for second cycle.
(c) P3
was loaded in increments up to 80.9 kips (load stage 26),
corresponding to factored derailment load (2.8 times the crush
30
(c) continued
live load and the design criterion (8),) and then unloaded to
find the residual deformations.
(d) P3 was loaded in increments to 131 kips (load stage 43-45),
corresponding to 28 in. deflection of the heavily-loaded stem.
The load was taken as the ultimate load which was 60% higher than
the required ultimate (factored) derailment load(= 4.5 times the
crush live load.)
In the derailment test, the loads, deflections, rotations and strand
strains were recorded after each increment. Crack pattern was traced on the
heavily-loaded stem on one end diaphragm at selected load stages.
Photographs were also taken for the final crack pattern. The maximum crack
widths at selected load stages were also recorded.
7.2 GIRDER 3
7.2.1 Torsional Fatigue Test
Before the torsional fatigue test, Girder 3 was first cracked by a
load system shown in Fig. 15b. The loads P1= P2 were applied by two MTS
actuators, while the two loads P5 was achieved by an auxiliary system of
cross heads and tie rods. The sequence of loading to crack the girder is
given in Table 6. The midspan deflectionswerealso measured by potention
meters. and were recorded in the Table. The pattemof cracking was photo
graphed and produced in Fig. 22A. The maximum crack length was about two
feet.
Girder 3 was subjected to 5.05 million cycles of loadings according
to Fisher's equivalent test Spectrum C, at a frequency of 4.0 cycles per
second. The test program was shown in Table 7. It can be seen that all
load cycles were 100% crush live load plus impact according to design
criteris (5). These were divided equally into load conditions I & II
31
7.2 continued
assuming that half of all the torques were positive and half were nega
tive. The fatigue test began and ended with load condition I, which is the
more severe case for the heavily-loaded stem.
The stimulated superimposed dead loads for Girder 3 should be P1
= 11.6 kips and P2 - 4.8 kips. These were adjusted slightly so that a
constant eccentricity could be maintained within each load condition.
7.2.2 Torsional Derailment Test
The derailment test program for Girder 3 is summarized in Table 8.
As for girder No. 2, loads P1 and P2 stimulating the factored dead loads
were first applied and then maintained constant throughout the test. The
derailment load P3
was then added in two steps:
(a) P3 was first loaded in small increments up to 58.2 kips (load
stages 16, 17; Table 8) corresponding to the unfactored derailment
load (=2 times the crush live load.) It was then unloaded to
find the residual deflections and rotations as a measure of the
serviceability of the girder after a severe derailment.
(b) P3 was applied in increments to 146.1 kips (load stage 47, Table
8) corresponding to 5.0 times the crush live load which represented
78.5% excess over the required (ultimate) factored derailment load.
In the derailment test, the loads, deflections, rotations and stirrup
strains were recorded after each increment. Crack patterns were traced on
the outside face of the heavily-loaded stem, on the heavily-loaded end
diaphragm and on midspan diaphragm. Photographs were taken for the final crack
pattern and the maximum crack widths were recorded at some selected load stages
8. TEST RESULTS AND ANALYSIS
8.1 GIRDER 1
8.1.1 Loss of Prestress
Loss of prestress in Girder 1 was given for the first 180 days in
Fig. 23. It could be seen that the measurements made by Demec and
Whittemore gages were very close. Loss of prestress increased very
rapidly up to 13% in the first three or four days and then slowed
down to approach the theoretical 23% loss asymptotically.
A comparsion of the experimental prestress loss with the theoreti
cal prediction was shown in Fig. 24. This figure indicated that the
experimental value was slightly higher than the theoretical value in
the early age and that this trend was reversed in the later age. In
general, however, the correlation between theory and test was good.
8.1.2 Camber Growth
The growth of camber for Girder 1 was plotted in Fig. 25. It
includes both the measured cambers and the theoretically predicted
cambers. It could be seen that the measured cambers were approximate
ly 7% less than the predicted cambers after about one month of age.
8.2 GIRDER 2
8.2.1 Natural Frequencies
The natural frequencies of the uncracked Girder 2 were measured
in flexure and in torsion before the initial static test. The girder
has been set-up on the supports as shown in Fig. 9 (without the load
ing systems and without tightening the hold-down rods at support).
Two LVDT's were installed at midspan near the north and south edges
32
of the flanges to monitor continuously the vertical deflection during
vibration. The vibration was started by a sudden impulse and was re
corded by a Hewlett-Packard strip chart recorder operating at a speed
of 2 inches per second. A sinusoidal curve was obtained, from which
the natural frequencies were determined.
The natural frequency in flexure was obtained by applying a
vertical impulse at midspan, so that the girder vibrated in a verti
cal direction. The natural frequency was measured to be 4.6 cycles
per second. This is very close to the calculated theoretical value
of 4.45 using the following equation
f = rEr / ---
2 .( 2 --.,: m
where E = modulus of elasticity of concrete (3,850 ksi)
I = transformed moment of inertia (813,731 in4)
m = mass per unit length= w/g, where w is the weight
per unit length (2.3 kips/ft.) and g is 32.2 ft/sec2 .
/- = span of beam (78.5 ft.)
The natural frequency in torsion was obtained by applying a
horizontal impulse at the bottom of the stem at midspan. This created
a torsional vibration about the shear center, which was picked up by
the vertical LVDT's. The horizontal vibration, however: was ignored
by the vertical LVDT's. In this way, the natural frequency
in torsion was found to be 9.2 cycles per second.
The theoretical natural frequency in torsion can be calculated
by the following equation, if the warping torsional resistance is
ignored:
33
where G = modulus of rigidity of concrete (1,610 ksi)
C = St. Venant torsional constant (76,235 in. 4)
i = moment of inertia of mass per unit length
= ( '(/g) I , where p ( is the unit weight (139 lb/cu.ft)
and I is the polar moment of inertia about shear p
center (254.4 ft.4)
J. = clear span between end diaphragms (75'-2").
This equation gives a theoretical natural frequency of 5.9 cycles
per second. Comparing this value with the measured frequency (9.2
cycles per second) shows that the warping torsional stiffness has
greatly increased the torsional natural frequency of the girder.
8.2.2 Initial Static Load
8.2.2.1 Vertical Deflections - Midspan deflections at the center
line were plotted as a function of truck loads from load stages
5 to 11 in Fig. 26. (See Table 2· for load stages). At the11e- load
stages of th e uncracked girder, the behavior was elastic.
A theoretical straight line was also included in Fig. 26.
This line was derived from the well-known flexural theory using
(a) the transformed section at mi dspan for the moment of inertia,
I and (b) the actual modulus of elasticity of concre t e, Ee,
measured from standard concrete cylinders (see Fig. 5). Ec was
taken as the secant modulus at the stress level of 0.1 f~. (see
Fig. 6). At this stress level, Ec = 0.76 (57,000 /iJ.), where a
34
reduction factor 0.76 was incorporated into the ACI Code expression.
Comparison of the theoretical str,aight line with the test data
showed excellent correlation. The elastic stiffness was found to be
101 kips/in. At a truck load of 72 kips, corresponding to crush
live load plus impact, the midspan deflection was 0.71 in.
A curve for loads vs. midspan deflections was also given for
cracked girder in Fig. 27. This curve, which was obtained from
load stages 55 to 61, remained straight (i.e. elastic). Fig. 27
showed that the flexural stiffness of cracked girder was 97 kips/in.
This was 4% lower than that for the uncracked section. At a truck
load of 72 kips the midspan deflection was 0.74 in.
Deflections along the girder axis were shown in Fig. 28 for
both the uncracked and the cracked sections. In view of symmetry,
the former was plotted on the left half-span while the latter on
the right. The experimental curves of all six load stages were in
cluded. Two load stages, namely 47.6 kips and 71.8 kips for the un
cracked section, were chosen to compare with the theoretical curves.
It could be seen that the correlation was excellent throughout the
girder.
8.2.2.2 Torsional Rotations-Midspan torsional rotation was plotted
as a function of truck torque from load stages 11 to 17 in Fig. 29.
At load stage 17, the girder was uncracked (see Table 2).
Three theoretical curves were also included in Fig. 29 for
(a) rigid end diaphragms (100% warping restraint), (b) no end
diaphragms (zero warping restraint), and (c) 22 in. thick end
diaphragms. 22 in. was the average thickness of a dapped end
diaphragm shown in the insert in Fig. 29. This simplification was
necessary because the computer program was written to treat only
35
uniform end diaphragms.(l, 5) In the analysis the span of girder
was taken as 75 ft. 2 in., the net span of the girder between the
inner faces of end diaphragms. The St.-Venant torsional constant,
C, and the warping moment of inertia, 1w,were assumed to be un
cracked and non-transformed. The modulus of rigidity was taken as
Ge = Ec/2(1 + p), where/' = 0. 2 and Ee = 0. 76 (57,000 /fb,).
The two theoretical curves for rigid end diaphragms and for
no end diaphragms provided the upper and lower limits for the test
eurves. The test curve, however, had a slope (torsional stiffness)
36
6% greater than the theoretical curve with 22 in. thick end diaphragms.
This small difference may be caused by the following three reasons:
(1) the average thickness of 22 in. may have underestimated the stiff
ness of the dapped end diaphragm, because the torsional stiffness
of the diaphragm is proportional to the third power of the thick-
ness. (2) the effect of reinforcement was neglected in assuming a
non-transformed section for St.-Venant torsional constant C, and
warping moment of inertia, 1w· (3) The Poisson's ratio}'-= 0.2 may
have been assumed too high.
The midspan rotations obtained from tests in Fig. 27 and the
corresponding differential deflections at midspan are given be low
foe the three design criteria (1), (2) and (3),
(1) Normal (2) Abnormal (3) Abnormal DESIGN CRITERIA Service Load Service Load, Service Load,
(see page 3) Tangent Track Tangent Track Curved Track
TORQUE/TRUCK (kip-ft.) 31 63 90
MIDSPAN ROTATION (deg.) 0.033 0.067 0.095
DIFF. DEFLECTION (in.) 0.035 0.070 0.099
37
The above table shows that when a girder of tangent track is subjected
to normal and abnormal service loads, the differential deflection at
midspan would be 0.035 in. and 0.070 in. respectively. These deflec
tions were less than 5% and 10%, respectively, of the deflections
under vertical service loads. Such negligible rotations due to
lateral loads should ensure the rider's comfort.
Loads vs. midspan rotation curve was also given for cracked
girder in Fig. 30. This curve was derived from load stages 61 to 67.
It showed that the torsional stiffness after cracking was identical
to that before cracking. Apparently the prestress had closed up
the flexural cracks and the girder behaved as though it had never
been cracked.
Torsional rotations along the girder were given in Fig. 31, for
both the uncracked and the cracked sections. The curves for the un
cracked section was plotted on the left half-span while those for
cracked section on the right half-span . Two load stages, namely
58.8 kip-ft. and 91.9 kip-ft. for the uncracked section, were chosen
to compare with the theoretical curves (22 in. diaphragms). The
correlation was quite satisfactory, but appeared to be better at 1/6
and 1/3 span than at midspan.
8.2.3 Flexural Fatigue Tests
8.2.3.1 Stiffness-During the fatigue test, the test was stopped
from time to time so that the static deflections could be measured.
It was found that the static deflection at the end of the test was
very close to that at the beginning. In other words, no apparent
deterioration of the stiffness was observed.
8.2.3.2 Cr acks -The crack wi dth of the largest crack was monitored
throughout the test. In the beginning of the fatigue test the crack
38
width was measured to be O. 001 1.n. under load condition I. This crack
widtb remained essentially the the same at the end of the fatigue
t~sts under the same load condition I.
8.2.3.3 Strains in Strands- The stress ranges in the prestressed
strands were measured during the last half of the fatigue tests for
load conditions I and IV and were recorded in the following Table.
Both these load conditions have (P1 + P2) min= 20.5 kips and (P1 + P2)
max= 92.5 kips. The only difference was the load eccentricity in
dicated in the Table.
LOAD CONDITION
IV
I
ECC.
1.5 in. NORTH
5.0 in. SOUTH
STEM
SOUTH
NORTH
SOUTH
NORTH
STRESS RANGE (ksi) * AVE
6.4-7.3
5.8-8.7
5.8-6.9
4.7-8.0
6.7
7.1
6.3
6 .... 1
*4 strand measurements were used in the south stem while ~lwere used
in the north stem.
A theoretical flexural analysis has also been carried out on Girder 2
to find the relationships among moment, curvature, steel strands from zero
load to failure. The tedious analysis utilized a trial and error method
to satisfy simultaneously the three conditions, namely equilibrium, com
patibility and stress-strain curves. The actual stress-strain curves of
prestressed strands and concrete were used. The stress-strain curve of
concrete was very closely approximated by a second-order parabola. This
analysis showed that under a load range from (P1 + P2) min= 20.5 kips
to{P. + P2) max= 92.5 kips, the stress range of the strands should be
7 .0 ksi.
The theoretical stress range of 7.0 ksi should be compared with
the measured stress range shown in the above table. It can be seen
that the measured stress ranges compared very well under load con
dition IV. The measured stress ranges were somewhat smaller than
the theoretical prediction, however, in load condition I. This dis
crepency was probably due to the difficulties of installing strain
gages on prestressed stands and the instability of the MTS actua
tors during this period of testing.
8.2.4 Flexural Derailment Test
39
8.2.4.1 Vertical Deflections - The midspan deflections at the center
line of girder 2 uere plotted as a function of axle loads in Figs. 32a
and 32b. Fig. 32a magnifies the strains and showed clearly the three
unloading cycles and their resulting residual deflections. The
measured curve should also be compared to the elastic theory for un
cracked section. Hence, the ratios of the measured deflection to the
uncracked theoretical deflection and the residual deflections were
given below for the three design load stages.
t-- --- -··- --------- -·-- --·--· ---- ------t ------------·-·------·-···---~ ---•··-•---·--··-·---·- ---·--· ·,---... ---· . !
Load
l I j i
Crush Live Load_ I Unfactored Derailment
(2 x Crush Live Load) Factored Derailment
(2.8 x Crush Live LoadD
Meas. Defl. Uncr. Theo. Defl.
1.06
1.46
2.17
Residual Deflection (in.)
0.09, 0.11
0.22
The above table shows that at a load of 2 times the crush live
load , corresponding to design criterion (7), the midspan deflection at
center line was only 46% greater than that predicted by the elastic
theory for uncracked section and the average residual deflection was
about 0.10 in.
Fig. 33 showed the deflections along the center-line axis. Due
to symmetry, only half of the span was measured for each load stage.
Four load stages were plotted namely, 1 times, 2 times, 2.8 times
and 3.8 times the cr ush live load. The measured deflections under
crush live load were compared with the theoretical curve for un
cracked section. The correlation was satisfactory throughout the
girder.
40
In a similar way, the midspan deflections of the heavily-loaded
stem were plotted as a function of axle loads in Figs. 34 and 35.
Fig. 34 showed that the residual deflections under 2 times the crush
live load and 2.8 times the crush live load were 0.14 in. and 0.36
in. respectively. Fig. 35 also showed that the test terminated when
the midspan deflection of the heavily - load stem reached 28 in. The
deflections along the heavily-loaded stem were recorded in Fig. 36
for the four load stages designated.
8.2.4.2 Torsional Rotations -The midspan rotations of Girder 2
wer e plot t ed as a function of axle torques in Figs. 37 and 38 . A
theore t i cal straight l ine based on uncracked section was also in
cluded in t hese figures. The r atios of the measured r otations to
the t heoretical r otati ons were given below for the three design load
stages. The r es idual rotations of the three unloading cycles were
also included.
Load Meas . Rotation Residual
Uncr. Theo. Rotation Rotation (deg . )
Crush Live Load 1.0 -Unfactored Derailment Loa< (2 x Crush Live Load) 1.49 0.045, 0.047 Factored Derailment Load (2.8 x Crush Live Load) 2.43 0.17
The above table showed that at a load of 2 times the crush live
load, corresponding to design criterion (7), the midspan torsional
rotation was only 49% greater than that predicted by theory for un
cracked section. The residual rotation at this load was only about
0.046 degree, corresponding to a differential stem deflection of
0.049 in •• Fig. 38 also showed that the test was terminated when
the midspan rotation reached 6.8 degrees.
Torsional rotations along the girder were recorded in Fig. 39.
Only half of the span was measured because of synnnetry. Four load
stages were plotted. The measured rotations under crush live load
were compared with the theoretical curve for uncracked section.
The correlation was found to be excellent.
In view of the small residual deflections and rotations under
severe derailment load, it can be concluded that the girders should
remain serviceable even after the derailment of vehicle, as defined
in the Load Criteria.
8.2.4.3 Ultimate Load - The ultimate load was determi ned from Figs.
32b and 35 to be 131 kips. When this load was r eached and maintained
constant, deflection and rotation continued to -increas e . The test
was discontinued when the maximum deflection under the heavily-
41
loaded stem was 28 in. and the maximum rotation was 6.8 degrees.
The termination was necessary, because (1) the gaps at the ends of
\ the girders between the bottom corner and the support blocks were
exhausted due to elongation of the bottom fibers, and (2) the
vertical tied rods for P1 was touching the inner face of the
heavily load stem and the tie rods for P3 was dangerously bent
above the slab on top of the cross heads. The large deformation
of the girder at ultimate load was shown in Fig. 14.
The ultimate load of P3 = 131 kips corresponds to 4.5 times
the crush live load. This load is 2.25 times the severe derail
ment load (design criterion (7)) and 1.61 times the factored derail
ment load (design criterion (8)).
8.2.4.4 Strains in Strands - The strains measured in the prestress
strands were plotted as a function of axle load P3 in Fig. 40. Those
measurements in the south stem (or heavily loaded stem) were indicated
by squares while those in the north stem by circles. These experi
mental measurements were compared to the theoretical curve based on
a sophiscated flexural analysis described in 8.2.3.3. It can be
seen that the trend of the experimental points is very well described
by the theoretical curve. Fig. 40 also showed that the strains in
the south stem were greater than the theoretical prediction, while
those in the north stem were less. This clearly showed the effect
of torsional moments in the derailment tests.
8.2.4.5 Crack Development - A general view of Girder 2 under derail
ment load has been given in Fig. 14. The cracking pattern of the
heavily-loaded stem were photographed along the girder in sections
and a composite picture was shown in Fig. 41. The process of crack
development in the derailment test was as follows: The vertical
42
flexural cracks was observed to have reopened at stage 6 when
P3 = 45.8 kips (see Table 5). The maximum crack width began to
develop and reached 0.008 in.at P3 = 58.9 kips (2 times crush
live load). It continued to increase to 0.019 in. when P3 =
80.9 kips (2.8 times crush live load). At load stage 25 (P3 = 67 . 4
kips) flexural shear cracks began to develop outside the load
points in the vicinity of the quarter points. This was followed
by the web shear cracks near the supports when the load reached
P3 = 82.6 kips.
It can be seen from Fig. 41 that all the cracks are well dis
tributed, resulting in small crack width. In terms of cracking the
girder behaved admirably in general. There was one crack at mid
span, however, that required examination. This crack began to open
very rapidly when P3 reached 106 kips and eventually caused the
failure of the girder. This phenomenon was apparently caused by
stress concentration of the midspan diaphragm and will be discussed
in the next section 8.2.4.6.
8.2.4.6 Midspan Diaphragm - Figs. 42a and b show an interesting
observation; i.e. the development of horizontal cracks in the stem
at midspan below the bottom face of the diaphragm. These horizontal
cracks started to develop rapidly at P3 = 106 kips and destroyed the
bond between the strands and the surrounding concrete. As a re
sult, the girder behaved like an unbonded girder in this region,
and one vertical flexural crack could widen very rapidly.
The cause of the horizontal cracks is explained in Fig. 43.
When one stem of a double-tee girder is subjected to a concentrated
load, part of this load must be transferred to the other stem by
43
the midspan diaphragm, In other words, the diaphiagm exerts an
upward shear on the stem heing loaded. This shear stress created
vertical tension just below the bottom face of the diaphragm and
produced the observed horizontal cracks.
No cracks were fotmd in the midspan diaphragm itself through
out the test. However, a fine crack was observed between the
diaphragm and the heavily loaded stem at high loads near ultimate.
8.2.4.7 End Diaphragms - The diagonal torsional cracking of one
end diaphragm is shown in Fig. 44. The cracking of the stem at
the jtmction with the inner face of an end diaphragm is shown in
Fig. 45. This typical cracking is produced by the differential
bending of the two stems (or warping torsion). Since these cracks
occurred at loads beyond the design derailment load (i.e. 2 times
the crush live load), they satisfy completely the serviceability
requirements.
8.2.4.8 Sectional Defonnation - No appreciable changes of the
cross sectional contourwereobserved until the load P3 exceeded
100 kips. This statement was true for both the cross sections
measured, namely, at quarter span and at 10 ft from midspan. At
load approaching failure, the change of the sectional contour
appeared to be irratic and therefore was not reported herein.
8.2.4.9 Construction Details - Cracks at the reentrant corners
44
started at load stage 25 (P3 = 67.4 kips). Since this is the same load
stage when flexural shear cracking occurred near the support, it appear-
ed that the reentrant corner did not cause any premature cracking.
The crack widths at the reentrant corners also remained small
throughout the derailment test. No distresses whatsoever were
observed at the hold-down points.
8.3 GIRDER 3
8.3.1 Torsional Fatigue Tests
8.3.1.1 Stiffness - Static deflections were taken from time to time
during the period of the fatigue test. No change of stiffness was
fotmd throughout the test.
8.3.1.2 Cracks Cracking of the girder prior to fatigue test pro-
duced flexural cracks in the stems (Fig. 22A) and one 5 in. long in
clined crack (about 25 deg. to horizontal) at the reentrant corner of
the <lapped end. All these cracks did not propagate during the
fatigue test.
45
8.3.1.3 Strains in Stirrups - The strains of two stirrups were moni
tored periodically throughout the fatigue test. The gages were placed
at No. 1 and No. 2 location on the outer leg of stirrup in the heavily
loaded stem. (see Fig. 19) Since these stirrups are located nearest
to the support and the gages are on the face where torsional stress
and shear stress are additive, the measured strains should be criti-
cal. The very small strains observed in these stirrup legs
during the test indicated that the concrete had not been cracked.
8.3.2 Torsional Derailment Test
8.3.2.1 Vertical Deflections - The midspan deflections at the center
line of Girder 3 was plotted as a ftmction of axle loads in Figs. 46
and 47. Fig. 46 show~ the unloading cycles after loading to 2 x the crush
live load (design criterion (7)) and the resulting residual deflec-
tions. The residual deflections were 0.070 in. immediately after un
loading and 0.049 in. after 1.3 hours. A theoretical curve based on
elastic analysis of uncracked section was also included in Figs. 46
and 47. This theoretical curve coincided with the measured curve
well beyond the crush live load. At 2 x the crush live load
~nfactored derailment load) the measured midspan deflection was only
5% greater than the theoretical value for uncracked girder.
Fig. 48 showed the deflections along the longitudinal center
line axis of Girder 3. Curves were plotted for three load levels at
1.0 times, 2.0 times and 2.8 times the crush live load. Comparison
of the test curves and the theoretical curves showed that the agree
ment was good throughout the girder at the first two load levels. It
should be pointed out that the curves were not symmetrical about mid
span.
In a similar way, the midspan deflections of the heavily-loaded
stem were plotted as a function of axle loads in Figs. 49 and 50.
Fig. 49 showed that the residual deflections under 2 times the crush
live load were 0.087 in. immediately after unloading and 0.057 in.
after 1.3 hours. The deflection,along the heavily-loaded stem were
also recorded in Fig. 51.
8.3.2.2 Torsional Rotations - The midspan rotations of Girder 3 were
plotted as a function of axle torques in Figs. 52 and 53. Fig. 52
gave the residual rotation of 0.024 in. after unloading from a load
of 2 times the crush live load. This residual rotation did not de
crease with the passage of 1.3 hours.
Figs. 52 and 53 also included a theoretical curve based on un
cracked section and 22 in. thick end diaphragms. It can be seen that
the measured torsional stiffness (slope of the curve) is 13% greater
than the theoretical torsional stiffness. Comparing this error with
the corresponding error of 6% in the initial static test of Girder 2
(Figs. 29 and 30), it appeared that the underestimation of the end
46
diaphragm stiffness discussed in 8.2.2.2 had a greater effect on
Girder 3 than on Girder 2. Apparently, the stiffness of the end
diaphragm had a greater strengthening affect on the torsional stiff
ness of the girder when the loads were applied close to the support
(Girder 3) than when the loads were symmetrical about midspan
(Girder 2).
The measured and theoretical torsional rotations along Girder 3
were plotted in Fig. 54, at three load levels. The unsymmetrical
shape of the curves was evident, especially at high loads.
8.3.2.3 Ultimate Load - The test was terminated when the load P3
reached 146 kips, corresponding to 5.0 times the crush live load.
This load is 2.5 times the severe derailment load (design criterion
(7)) and 1.78 times the factored derailment load (design criteria
(8)). At this load level the midspan deflection of the heavily
loaded stem was 7.5 in. (Fig. 50) and the midspan rotation was 2.8
degrees (Fig. 53). The stirrups in the most highly stressed region
have also yielded. Figs. 59 to 61 showed the cracking condition at
this load level. It should be noted that the ultimate capacity of
Girder 3 has not yet been reached.
8.3.2.4 Strains in Stirrups The stresses in the stirrups were
sunnnarized in Figs. 55 to 58. The location of the gages were given
in Fig. 19. Fig. 55 gave the stresses in the outer legs of stirrups
in the heavily-loaded south stem. It can be seen that three of the
gages close to the support have yielded before termination of the
test. The other five gages, further away from the support, have
substantial stresses but did not yield.
47
Fig. 56 gave the stresses in the inside legs of stirrups in
the heavily loaded south stem. The stresses in these stirrups were
substantially less than the yield stress since the stresses due to
shear and the stresses due to torsion are subtractive.
Fig. 57 showed the stresses in the outer legs of stirrups in
the north stem. The stresses in these gages were very small. In
other words, the shear and torsional capacity of this stem _had bare
ly been utilized.
Fig. 58 gave the stirrup stresses in the end diaphragm. The
average stress was about 20 ksi, which was about one-third of the
yield stress.
8.3.2.5 Crack Development - The previously induced vertical flexural
cracks in the heavily-loaded south stem was observed to reopen at a
load of P3 = 51.3 kips. These cracks developed very slowly. The
maximlll!1 crack widths was only 0.003 in. at P3 = 58.2 kips (2 times
crush live load), and became 0.010 in. at P3 = 108. 7 kips. Fig. 59
showed the flexural cracking pattern at 5.0 times the crush live
load (2.5 times the severe derailment load). The cracks were well
distributed and finely spaced.
The flexural shear crack started to spread at the high factor
ed derailment load of P3 = 80.9 kips (2.8 times crush live load).
These cracks first occurred between the innermost load point and
the midspan, and then spread toward the east support. Most of
these shear cracks lied within the hold-down points where the stir
rup spacing was 20 in. The crack widths of these shear cracks
developed very rapidly, reaching 0.050 in. at P3 = 108.7 kips and
0.090 in. at P3 = 139.6 kips. The cracking pattern of these shear
48
cracks in the vicinity of midspan was given in Fig. 60.
8.3.2.6 Re-entrant Comer of Dapped End - The pre-existing small
crack (see Section 5.2.1) at the re-entrant comer of the <lapped end
first re-opened at a low load of P3 = 14.4 kips with a maximum
crack width of 0.002 in .• This maximum crack width, however, in
creases very slowly and reached only 0.005 in. at the derailment
load of P3 = 58.2 kips. It was 0.011 in. at P3 = 108.7 kips and
0.017 in. at P3 = 140 kips. The cracking patterns are shown in
Fig. 61a at P3 = 58.2 kips (2 times crush live load) and in Fig.
61b at P3 = 146.1 kips (5 times cruch live load).
8.3.2.7 Joint Between Stem and End Diaphragm - The joint crack
between the stem and the end diaphragm was first observed at
P3 = 51.3 kips. This crack developed rapidly, and the crack width
was recorded to be 0.010 in. at the derailment load of P3 = 58.2
kips. At higher loading this crack continued to widen, reaching
1/8 in. at P3 = 110 kips. This was accompanied by spalling of con
crete and longitudinal splitting at the bottom of the stem. Figs.
62a and 62b showed the condition of this joint at 5 times crush live
load or 2.5 times severe derailment load (P3 = 146 kips).
8.3.2.8 End Diaphragm - The pattern of diagonal t orsional cracking
at the west end diaphragm is shown in Fig. 63 following the termina
tion of test (P3 = 146.1 kips). These cracks developed very slowly.
The maximum crack width was 0.00 8 in. at the severe derailment load
of P3 = 58.2 kips . They became 0.015 in. at P3 = 108.7 kips and
0.040 in. at P3 = 139.6 kips.
49
8.3.2.9 Midspan Diaphragm - Midspan diaphragm was observed to develop
torsional cracks only at th e very high load of P3 = 146.1 kips . This
was 5.0 times the crush live load. The cracking pattern on both
sides of the diaphragm is shown in Fi~. 64.
It should be noted that the local horizontal cracks in the
stern at midspan, which was very prominent in Girder 2, (Figs. 42a
and 42b) also existed in Girder 3. These local horizontal cracks
are shown in Fig. 65. Their widths, however, were small.
8.3.2.10 Sectional Deformation - No appreciable deformation of
the cross-sectional contour was observed until P3 reached 80.9 kips.
At the termination of the test at P3 = 146.1 kips, the contour
deformation was still very small.
50
9. BENDING AND PUNCHING TEST OF FLANGE SLAB
9.1 OBJECTIVES
An additional test was made on Girder 3 to study the bending and
punching behavior of the overhanging flange slab under derailment con
dition. Two wheel loads, 7 ft.-6in. apart, were assumed to be located
at a distance 8-1/4 in. from the edge of the flange, just inside the
7 in. wide curb. The bearing area of each wheel was assumed to be 2 in.
by 8 in •.
The purpose of this test was three-fold:
(1) To check the bending or punching capacity of the flange
slab and the behavior of the flange under derailment
design criteria (7) and (8).
(2) To check the validity of two equations proposed by Tudor
Engineering Co. for the effective width E, over which a
wheel load could be dispersed at the face of support of
the overhanging flange. These two equations are:
E = 0. 85 x + 5 . 5 ' ~ 7. 5 '
E = 1.4 x + 2.2' ~ 7.5 1
(9 .1)
(9. 2)
where xis the distance in feet from the load to the face
of supports and 1. 75 '< x < 3.0'. The top bars ( No. 5
at 8 in. spacing) in the overhanging flange of the test
girders were calculated on the basis of Eq. (9.1). Using
Eq. (9.2), however, the top bars would have to be in
creased to No. 6 at 8 in. spacing.
(3) To check the adequacy of the longituoinal distribution re
inforcement provided in the overhanging flange.
51
9.2 TEST SET-UP AND INSTRUMENTATION
Two loading points, P4, were established at sections G and H as shown
in Fig. 66a. These load points, each representing one wheel load, were
7 ft. - 6 in. apart and were located 8-1/4 in. from the edge of the
flange (Figs. 66b and 66c). The load ·at each point was transmitted to
the slab through a 2 in. x 3 in. x 8 in. steel block with a bearing sur
face of 2 in. by 8 in •• The relative position of the steel block with
respect to the curb and the reinforcement was given in Fig. 66d.
Each load was applied through a system of crossheads and tie
rods (Fig. 67a) and was activated by a ram. The oil pressure for the
two rams was supplied from the same pump to provide simultaneous loading
of equal pressure. Each of the two loads was measured by a 50-ton pre
calibrated load cell. It was found that the two loads may differ as
much as 11% due to friction in the rams and the hoses.
Deflections of the flange were measured by potentiometers each
mounted on a steel tube connected rigidly to both stems of the girder
as shown in Fig. 67b. A potentiometer was provided for each load point
at approximately 5 in. from the center of the load point on the straight
line connecting the two load points.
The loads and the deflections were recorded continuously by two
x-y plotters.
9.3 TEST PROGRAM
Each wheel load P4 was increased in increment with the target values
of 5, 10, 15, 20, 25 and 28.9 kips. The last value was two times the
crush live load according to the severe derailment criterion (7). The
flange was then unloaded to find the residual defle ction. Followi ng
52
an hour of rest, the flange was reloaded in three steps to a target
-value of 28.9 kips.(severe Derailment .Load, Criterion 7). The load
was further increased with small increments of 3 kips to 40.5 kips, or
2.8 times crush live load according to factored derailment criterion (8).
Finally, the load was increased to failure.
9.4 TEST RESULTS
The test results were summarized in Table 9. The load-deflection
relationship was plotted in Fig. 68 using the average value of the two
loads and the average value of the two deflections. At a load P4 =
20.9 kips, the very fine pre-existing cracks caused by torsional derail
ment test on top of the flange reopened. At a load P4 = 26.0 kips new
cracks on top of the flange began to develop in roughly concentric
circles around the two loads. These cracks multiplied and widened to
53
a maximum crack width of 0.005 in. and an average crack width of 0.004 in.
at P4 = 29.9 kips. This load corresponded approximately to 2.0 times
the crush live load. After unloading, the residual deflection was
about 20% and the maximum crack width was less than 0.001 in.
Reloading of the flange showed that the load-deflection relation
ship was quite linear up to P4 = 29.7 kips (see Fig. 68). At this load
crack widths were also the same as those before unloading. When the
load was increased to P4 41.7 kips, corresponding approximately to
2.8 times the crush live load, the average crack width became 0.010 in.
At P4 = 44.6 kips cracks began to open on the bottom face of the
flange. This crack developed inward about half way to the stem. Failure
occurred abruptly and loudly by punching shear around load point H,
at an average P4 value of 47.1 kips. This failure load was 3.26 times
the crush live load or 1.63 times the derailment load. After failure
at load point H, load point G was loaded individually until failure at
P4 = 51.3 kips. Failure was again due to punching shear. It was in
teresting to note that some ductility was observed before this abrupt
failure. This apparent ductility might be caused by the deformation
around the bearing area or might be due to the yielding of reinforce
ment in the slab. Fig. 69 showed the cracking pattern, and Figs. 70
and 71 showed the mode of failure at load pointSH and G. respectively.
9.5 SUMMARY STATEMENTS
(1) The flange was able to resist 1.6 times the severe derail
ment load (3.2 times crush live load). This capacity ex
ceeded the design requirement of 1.4 times the severe de
railment load. At the derailment load the cracking con
dition and the residual deflection were tolerable.
(2) Although the flange slab did not fail in bending, it is
possible to state that Eq. (9.1) for the effective width
is conservative. Consequently, the top bars of No. 5 at
8 in. spacing is adequate to satisfy the design criterion.
(3) The longitudinal distribution reinforcement appears to be
satisfactory for the girder. Additional reinforcement is
not expected to improve the behavior of the overhanging
flange.
(4) The pre-existing cracks introduced during the derailment
test of Girder 3 had very little effect on the results
of the test.
54
10. CONCLUSIONS
1. The girders behave elastically both in flexure and in torsion well
beyond the abnormal service load of 72 kips vertical load plus 63
kip-ft. torque per truck. At this load, the maximum midspan deflec
tion was 0.71 in. and the maximum midspan rotation was 0.067 deg.,
corresponding to a differential stem deflection of 0.070 in. These
small deformation should ensure rider comfort.
2. Introduction of flexural cracks due to accidental overload or
environmental factors (secondary stresses) does not change the
elastic behavior of the girders, both torsionally and flexurally,
well beyond the abnormal service load level.
3. The flexural behavior of the girders can be accurately predicted
by the well-known flexural theory, both before and after cracking.
In this calculation the modulus of elasticity is taken as:
EC= 0.76 (57,000 )tI)
The reduction factor 0.76 reflects the actual property of the
Florida oolite concrete.
4. Torsional behavior of the girders in the elastic range can be reason
ably predicted by Vlasov's elastic mixed torsion analysis, taking
into account both the St.-Venant torsional resistance and the warp
ing torsional resistance.
5. The natura l frequencies of the girders are 4.6 cycles per second
in flexure and 9.2 cycles per second in torsion. These natural
frequencies are much higher than those of the vehicles.
6. Girder 2 and Girder 3 have been subjected to different test spectra of
55
6 million cycles and 5 million cycles, respectively, of dynamic loading,
equivalent t o 60 years of service life. No deteriorations whatsoever
have been observed in the girder in terms of flexural and torsional
stiffnesses, crack propagation and stresses in the strands.
7. Under severe derailment load (crush live load with 100% impact and
a maximum sideshift of 3 ft.), the maximum midspan deflection- and
rotation were 1. 72 in. and 0.61 degree, respectively. These de
formations were only 46% and 49%, respectively, greater than those
predicted by the elastic theory. The residual deformations were
also very small (0.11 in. for deflection and 0.047 degree for
rotation). In other words, the girders remained serviceable after
severe derailment.
8. The ultimate load of Girder 2 was 2.25 times the severe derail
ment load, or 1.61 times the factored derailment load (load factor
of 1.4)
9. The stirrups were designed by the elastic mixed torsion analysis
tmder abnormal service load and by the "post-cracking mix torsion
analysis" tmder factored derailment load. This resulted in No. 4
bars at 10 in. spacing between the hold-down points and the points
3' - 3½" from support. This design was fotmd by tests to be very
conservative.
10. Prestress losses and cambers can be predicted to a reasonable
accuracy. The small camber growth after installation of the rails
is expected to improve rider comfort.
11. Cracks were very few in number and crack width very small at the
re-entrant corners of the <lapped ends, and at the end diaphragms
under abnormal service load. No distresses whatsoever had been
observed at the hold-down points.
12. The midspan diaphragm appears to have introduced local stress con-
56
centration in the stem resulting in local horizontal cracks and the
opening of one large crack near ultimate. Since this process
occurred at very high load near failure (1.9 times the severe de
railment load), it did not detract from the generally excellent
behavior of the girder.
13. It is reconnnended that the center diaphragm of the girder be re
tained for the Stage I construction of the Dade County Rapid Transit
aerial guideway, since it may have contributed to the ultimate
strength of the girder. However, future research is desirable to
find out whether the midspan diaphragm can be eliminated for the
Stage II construction.
14. The overhanging flange is able to resist 1.6 times the severe de
railment load (3.2 times crush live load). The reinforcement pro
vided in the flange of the test girder appears to be adequate.
15. In view of the observed separation between stem and end diaphragm
at the joint, when the girder is subjected to derailment loading,
it is recommended that all main reinforcement from the end diaphragm
be extended into the stems with full anchorage.
57
58
11. RECOMMENDATIONS
The excellent behavior of the tested girders both in service
ability and in strength substantiated all the analysis and design
methods used. The tests demonstrated conclusively that double-tee
girders are viable standard girders for the aerial guideways of mass
rapid transit systems. It is reconnnended, therefore, that this pre
stressed double-tee girder be an approved alternative for the aerial
guideway for the Stage I construction of the Dade County Rapid Transit
System.
12. REFERENCES
1. Hsu, T. T. C., "Torsion Analysis of Dade Cotmty Rapid Transit Aerial Guideways", Report to Kaiser Transit Group, April 12, 1977.
2. Hsu, T. T. C., "Torsion Tests of Double-Tee Girders for Dade County Rapid Transit Aerial Guideways", Report to Kaiser Transit Group, August 23, 1977.
3. PCI Design Handbook, Prestressed Concrete Institute, First Edition, 1971. Second Edition, 1978.
4. Dhondy, S. and Jaurequi, I., "Detailed Calculations of 79 ft. Effective Span Double-Tee Girders for Tangent Double Track". Internal Report of Kaiser Transit Group, Aug. 1977.
5. Hsu, T. T. C., "Computerization of Flexure and Torsion Design for Dade County Rapid Transit Aerial Guideways". Report to Kaiser Transit Group, May 29, 1978.
6. Vlasov, V. Z., "Thin Walled Elastic Beams", 2nd Edition, Moscow, 1958. Translated into English in 1961, available from the National Technical Information Service, U.S. _Department of Commerce, Springfield, Virginia.
7. Zia, P. and McGee, D., "Torsion Design of Pres tressed Concrete", Journal of the Prestressed Concrete Institute, Vol. 19, No. 2, MarchApril, 1974.
8. Zia, P. and Hsu, T. T. C., "Design for Torsion and Shear in Pres tressed Concrete". Preprint 3424 ASCE Fall Meeting, Chicago, Oct. 16-20, 1978.
9. Givens, W., "Review of the Design and Producibility of Double-Tee Test Girders for Stage I - Rapid Transit System Metropolitan Dade County Transportation Improvement Program. Final Report to Kaiser Transit Group, Feb. 1979.
10. Collins, J. and Hsu, T. T. C., "Properties of Florida-Oolite Concrete1'.
Journal of Testing and Evaluation. ASTM, Vol. 5, No. 3, March 1977.
11. Hanson, N. W., Hsu, T.T.C., Kurvits, 0. A. and Mattock, A.H., Facilities and Test Methods of PCA Structural Laboraoty - Improvements 1960-65", PCA Development Department Bulletin, D 91, 28 pp. 1965.
12. Fisher, J. W., Hwang, T. and Slutter, R .. G., "Selection of A Fatigue Load Test Spectrum for a Pretensioned Concrete Double-Tee Girder". Report to Kaiser Transit Group, March, 1978.
59
60 13. METRIC CONVERSION TABLE
MEASUREMENT ENGLISH UNIT METRIC UNIT
LENGTH 1 inch (in. or II) 2.540 centimeter (cm)
1 foot (ft. or ') 0.3048 meter (m)
1 yard (yd.) 0.9144 meter (m)
1 mile (mi.) 1.609 kilometer (km)
AREA 1 sq. inch (in. 2) 6.452 sq. centimeter (cm2)
1 sq. foot (ft. 2) 0.09290 sq. meter (m2)
1 sq. yard (yd.2) 0.8361 sq. meter (m2)
1 sq. mile (mi2) 2.589 sq. kilometer (km2)
VOLUME OR (in. 3) or cm3 STATIC MOMENT 1 cu. inch 16.39 milliliter (ml)
1 cu. foot (ft.3) 28.32 liter (1)
1 cu. yard (yd. 3) 0.7646 cu. meter (m3)
1 ounce (oz.), U.S. Fluid 29.57 milliliter (ml) or cm3
1 gallon (gal.) 3.785 liter (1) --MOMENT OF
INERTIA 1 in.4 41.62 cm4
1 ft. 4 0.008631 m4 . -
SECTORIAL MOMENT 1 in. 6 268.5 cm6 OF INERTIA
1 ft. 6 6 0.0008018 m ··-
FORCE 1 pound (lb.) 4.448 newton (N) .
0.4536 kilogram force (kg-f)
1 kilopound (kip or k) 4.448 kilonewton (kN)
1 ton (200 lb.) 0.9072 metric ton (1000 kg)
!DISTRIBUTED 1 lb./ft. (plf) 14.59 N/m FORCE 1 kip/ft. (klf) 14.59 kN/m
1 lb./in. (pli) 175.1 N/m
STRESS 1 lb./in. 2 (psi) 6.895 kN/m2
1 kip/in. 2 (ksi) 6.895 MN/m2
1 lb./ft. 2 (psf) 47.88 N/m2
1 kip/ft. 2 (ksf) 47.88 kN/m2
DENSITY 1 lb./ft. 3 0.1571 kN/m3 -~ MOMENT 1 lb.-in. 0.1130 N-m or 1.152 kg-cm
1 lb.-ft. 1.356 N-m or 13.83 kg-cm
1 kip-in (k-in.) 0.1130 kN-m or 11.52 kg-m
1 kiE-ft . (k-ft.) 1.356 kM~m or 138.3 kg-m ---
TEMPERA'.l'URE 1°Fahrenheit (oF) (oF-32)/1.8 Celsius (oc)
Table 1. LOCATIONS OF VERTICAL TRANSLATION MEASUREMENTS
GIRDEE TEST SECTION POINT
WEST DIAPH 2 4
INITIAL A-A 1, 2, 3, 4, 5
STATIC B-B 1, 2, 3, 4, 5 2 AND
DERAILMENT c-c 1, 2, 3, 4, 5
EAST DIAPH 2 4
l
WEST DIAPH 3
FATIGUE c-c 1 3 5
EAST DIAPH 3
WEST DIAPH 2 4 I
A-A ! 2 4 i
B-B I 2 4 )ERAILMENT ,j c-c I 2, 3, 4
3 F-F l 2 4
EAST DIAPH 2 4
I I WEST DIAPH I 2
l
I i;'ATIGUE c-c l 2 4 I
i EAST DIAPH ' 2
I l l I I l
i I
TABLE 2 INITIAL STATIC TEST PROGRAM - GIRDER 2
LOAD I . . ~,.
LOAD UNCRACKED LOAD CRACK CRACKED STAGE STATIC TEST ~ STAGE GIRDER STAGE I STATIC TEST ~
pl . p2 T ~ pl p2 pl p2 I T: ~ ~ I I ~
(k) (k) (k-ft) (k) (k) (k) (k) ! (k-f~)
i ! i I
/31 .
I 1,2 0 ,..:i 27 0 0 52 0 0 i i,..; 0 l 0
3 2.5 CJ) 28 46 46 53 4.3 2.5 I . CJ)
:>-t :>-t 4 5.1 ,..:i 29* 48 48 54 8.6 5.1 i 8.6 1 p., ,..:i
p., I p., 5 12.9 7.6 < 30 50 50 55 12.9 7 .6 , I~
l ,,...._
6 18.9 . 13.6 H 31 52 52 56 18.9 13.6 ! H I H~ H~
7 24.9 I 19.6 f2 ffi 32 53 53 57 24.9 19.6 CJ) ::.:: w w HP:: Hp:::
8 30.9 25.6 u 33 54 54 58 30.9 25.6 u ...:iz ~ ~
9 36.9 31.6 ~H
34 55 55 59 36.9 31.6 ::::> CJ) ::::> CJ)
fJ e: >::: p., 10 42.9137.6 35** 56 56 60 42.9 37 .6 ! WH
,..:i ::,:: ~::,:: i .... 11 48.9 43.6 0 '° 36 57 57 61 48.9 43.6 ! 0 ~ ,.__, _ .,.,_., __ , __ ...,__ __ , _ __ ,.. _
....._..._ ... . ,-- -·~ ,.....__ .... -~ ..... ·-~+-- ---r-··-. 12 .'.>1.9 40.6 15
,,...._ 37 58 58 62 51.9 40.6 1 15 -;:;-H
~ ' H~
' 13 54.9 37.6 30 H ;:.:! 38 59 59 63 54.9 37.6 ; 30 CJ) w l ~~ 14 5 7. 9 34.6 45 w p::: 39 60 60 64 5 7.9 34.6 I 45 HU I HU z~ z
15 60.9 31.6 60 40 61 61 65 60.9 31.6 I 60 ZH ' 0 0 H~ ~ ~ 16 63.9 28.6 75 ~I 41 62 62 66 63. 9 28.6 75 p::: I 0~ 0~ H tn H Lr) 17 66.9 25.6 90 42 63 63 67 66.9 :25.6 90 ,-I 1;;- I 18 60.9 31.6 60 43 64 64 68 60.9 , 31.6 60 OW
19 54.9 37 . 6 30 < ::::> 44 65 65 69 30 3g 54.9 37.6 0 O'
20
21
22
23
24
25
26
48.9 43.6 0 s~ 45 66 66 70 48.9 0 46 67 67 71 36.9 36.9 31.6 <
~3 24.9 19.6 47 0 0 72 24.9 0 ,..:i~
12.9 7.6 s~ 48 0 0 73 12.9 · ~
8.6 5.1 49,50**" 67 67 74 8.6 0 .:t:...:i 4.3 2.5 0 51 0 0 75 4.3 ...:io es~ ' 0 0 76 0
*Cracking at stress raiser; P1=P2=47.5k; Max. crack width <0.001"
**Cracking in flexural region; P 1 =P 2=55. 7k; Max. crack width -=- 0 .002"
k k ***Crack length reach 1. 5 ft; P =6 7. 5 , P =66. 0 ; Max. crack width = 0. 006" 1 2
,..:i p:::
43.6 0 s~ 31.6
0 O<t: <o
19.6 o...:i ,..:i . s~
7.6 ~
5.1 0
2.5 <...:i 0 ...:io
0 is~
I I
r i
I I
I
r
TABLE 3
I LOAD l I•
TARGET FATIGUE TEST PROGRAM - GIRDER 2
(6-MILLION CYCLE)
; 1 i
i LOAD pl p2 CYCLE ! CONDITION I (MILLIONS) ! (KIP) (KIP) CLL+I
l I ! I I t 12.01 8.51
I 0.800 I 53.9 38.6 100%2 I I
I 12.0 8.5 II 0. 700 49.4 35.1 89.1% ! ! 9.81 10.71
III 1.400 ! ( 40.4 44.1 89 .1% t t I I i
j
9.71 10.81 I IV 1.600 ( l
I
' 43.9 48.6 I 100% I I
l 12.0 8.5 i
II 0.700 • l 49.4 35.1 . 89.1% i
I 0.800 12.0 8.5 ' 53.9 38.6 100%
6.000
1superimposed Dead Loads should be simulated by P1 = 12.9k,
REMARK
LOAD TORQUE ECC. (IN.)
+(N+W) 3 5.0 S
+(N+W) 4 5.0 S
-(N+W) 1.3 N
-(N+W) I.SN
+(N+W) 5.0 S
+(N+W) 5.0 S
k Pz = 7. 6 , but
were changed to maintain a constant load eccentricity indicated.
2cLL + 25% impact gives P1 = P2 = 36k
310% nosing+ 13.5 MPH wind= z5k-ft
410% nosing+ 13.5 MPH wind= 22.5k-ft
j I i j
l )
;
TABLE 4. ACTUAL FATIQUE TEST PROGRAM - GI:RDER 2
LOAD REMARK LOAD
CYCLES pl p2 CLL+I TORQUE LOAD
CONDITION (MILLIONS) (KIP) (KIP) ECC. (IN.)
I 0.610 12.0 8.5 53.9 38.6 100% +(N+W) 5.0 S
II \ 1.081 12.0 8.5 I l 49.4 35.1 89.1% +(N+W) 5.0 S I
I
III 1.196 9.8 10. 7
40.4 44.1 89.1% -(N+W) 1.3N
IV 1.632 9.7 10.8 \ 43.9 48.6 100% -(N+W) 1.5N ' I
I ' 1
I I
l 9.7 10.8 III* 0.200 I
' I I 40.4 44.7 89 .1% -(N+W) 1.5N I
I
i
i 12.0 8.5 I II 0.311 t I 49.4 35.1 ; 89 .1% +(N+W) 5.0 S ! '
I 0.990 ) 12.0 8.5
I l
I ) 53.4 38.6 100% +(N+W) 5.0 S
! i
6,020
* An eccentricity of -1.5 in. was used instead of the -1.3 in •. This was done to avoid the changing of actuator position and should be conservative.
I
TABLE 5. DERAILMENT TEST PROGRAM - GIRDER 2
P 3 (kips) I
LOAD REMARK STAGE TARGET ACTUAL
(AV.)
'
1-3* 0 0 Apply P1 ,P2
4 14.4 12. 3
5 28. 9 27. 91 Crush L. L.
6-8 43.3 45.8
9 ,10 57.8 58. 92 2.0xCrush L. L.
11 28.9 31. 7
12-15 0 0
16 14.4 11.9
17 28. 9 27. 8
18 43.3 44.2 I
19,20 57. 8 57.7 I 21 0 0
I ! I
Simulated Superimposed Dead Load P1 = 7.9k P2 = 15.4k
¾fax. crack width -::0.001 in. (cracks did not reopen)
2Max crack width= 0.008 in
LOAD P
3 (kips)
REMARK STAGE TARGET ACTUAL
(AV.) '
22 0
23 57. 8 56.7
24 63.5 62.3
25 69.3 6 7. 4 F.S. Crack
26 80 .9 80.9 3 2. 8xCrush L. L.
27 0
28 80 .9 79. 3
29-30 82.6 W. S. Crack
31 93. 8
32-34 106 4 3. 7xCrush L. L.
35 110 3.8xCrush L. L.
36-39 120
40-42 127
43-45 131 4. 5xCrush L. L.
3Max. crack width• 0.019 in. 4Max. crack width= 0.187 in. (one crack open at
midspan)
TABLE 6. LOADING TO CRACK GIRDER 3
* LOAD (P1+P2), kips (P
1+P2), kips * 2P5 , kips MIDSPAN
STAGE at 6'-3" ** at 2.7 '-3" at 30 '-3" DEFLECTION ( IN.)
1,2 0 0 0 0
3 9.8 10.0 0 0.057
4 20.0 20.6 0 0.115
5 30 .1 30.6 0 0 .170
6 40.2 40.7 0 0.227
7 50.4 50.9 0 0.287
8 60 .5 60.5 0 0.343
9 70 .6 70. 4 0 0 .401
10 80 .6 81.8 0 0.465
11 90.4 91.4 0 0.525
12 95.5 97. 3 0 0.559
13 95.4 102.1 0 0.590
14 95.6 101.0 7.2 0.625
15 95.3 101.3 15.4 0.666
16 95.7 100.9 26.3 0.720
17 95. 7 101.1 39. 4 o. 790
18 95.7 100 .4 53. 2 0. 858
19 95.7 101.4 66.3 0 .936
20 95.6 100.9 80. 4 1.015
21 95.8 100.4 86.9 1.059
22 95.8 101.5 89. 4 1.090
23 95.8 100.3 91. 8 1.100
24 95.6 100.7 95.8 1.132
25 95.6 101.9 95.9 1.157
* See Fig. 15b, P1 = P2 .
** Distance measured from center of support.
TABLE 7. FATIGUE TEST PROGRAM - GIRDER 3
I .
LOAD LOAD CYCLES pl p2 REMARK (MILLION) LOAD :
CONDITION TARGET ACTUAL (kip) (kip) CLL+I TORQUE ECC. (IN.)
I 10.61 5. a1 I 1. 2625 1.352 i 57.3 31.1 100%
2 (N+w+E+c. F.) 3 8. 9 S
8. s1 7.61
(-N~W+E+C.F.) 4 II 2.525 2.450 47.3 41.1 100% 2.1 S
10.6 5.8 I 1.2625 1.247 l 57.3 31.1 100% (N+w+E+C. F.) 8.9 S
I 5.050 5.049
1superimposed dead loads should be simulated by P1 = 11.6 kips, P2 = 4.8 kips
but were changed to maintain a constant load eccentricity indicated.
2CLL + 25% impact gives P1 = P2 = 36 kips
310% nosing+ 13.5 mph wind+ Ecc. for 5200 ft. radius+ C.F. for 70 mph= 48.5k-ft.
~10% nosing - 13.5 mph wind+ Ecc. for 5200 ft. radius+ C.F. for 70 mph= l.5k-ft.
!
TABLE 8 DERAILMENT TEST PROGRAM - GIRDER 3
LOAD p3 LOAD p3 STAGE (Kips) REMARK STAGE (Kips) REMARK
1-3 0 Apply pl' p2 1 25 21.9
4.5 5.8 26 29.3 CLL
6 7.9 27 33.6
7 10.1 28 44.7
8 12.3 29 58.1 2.0xCLL 3
9 14.4 Reentry corner 30 63.9 crack reopen .002"
10 17.8 31 69.7
11 25.0 32 76.1
12 29.0 CLL (28.9) 33,34 80.9 2.8xCLL (80.9)
13 33.6 35 89.8
14 43.9 36 98.9
15 51. 3 Joint crack2 open 37 100.6 Flex. cracks reopen
16,17 58.2 2.0xCLL (57.8) 3 38,39 108.7 3.8xCLL (109.8) 3
End Diaph. cracks i
18 30.8 ; 40 117.8 l
19,20 0 I 41 129.2
21 3.2 I
42 131.2 l
5.9 !
43 137 .6 22
I 23 11.7 44,46 139.6 4.8 CLL3 I
24 18.1 I 47 146.1 5.0 CLL (144.5)
! Midspan diaph.
t cracks.
1simulated superimposed dead loads P1 = 6.6 kips, P2 = 12.8 kips. 2Joint crack is crack between the stem and the end diaphragm.
3Max. crack widths are (in inches):
58.1
108.7
139.6
FLEXURE
. 002
.010
.015
FLEXURAL SHEAR
.050
.090
REENTRY CORNER
. 005
.011
.017
END DIAPH.
.008
.015
. 040
.010
.120
.500
TABLE 9. RESULTS OF FLANGE SLAB TEST.
LOAD P 4 (KIPS) DEFLECTIONS (in.) STAGE REMARK LOAD G* H* AV. G H AV.
0 I 0 0 0 0 0 0
1 5.5 5.0 5.3 .007 .005 .006
2 11.0 9.8 10.4 .016 .013 .015
3,4 16.1 15.0 15.6 .027 .025 .026
5 21.5 20.2 20.9 .045 .043 .044 Existing Cracks Open
6 26.8 25.2 26.0 .068 .067 .068 New Cracks Open
7 30.8 29.0 29.9 .086 .086 .086 2.0 x CLL (P4=28.9k), wmax = .005"
8 0 0 0 .018 .017 .018 20% Res. Def 1. , Wmax = • 001"
9 0 0 0 0 0 0
10 11.0 10.0 10.5 .022 .022 .022
11 22.0 20.0 21.0 .048 .046 .047
12 31.1 28.3 29.7 .072 .070 .071 k " 2.0 x CLL (P4=28.9) Wmax = .005
13 34.1 31.2 32.7 .083 .080 .082
14 37.7 34.3 36.0 .103 .100 .102
15 41.0 37.2 39.1 .116 .113 .11.S
16 43.6 39.8 41.7 .133 .134 .134 2.8 x CLL (P4=40.4k) Wmax = .010"
17 46.6 42.6 44.6 .143 .145 .144 Cracks Open On Underside of Flange
18 49.1 45.0 41.1 - FAILED - 3.26 x CLL, (P4=47.lk)
19 0 0 0 .019 - -20 51.3 0 - FAILED - - - k) 3.55 x CLL, (P 4-51.3
*Location of Wheel Load (see Figs. 66a and 66b)
,.
1'2' -o• c.·-o··
2'- c.' _I_ ?>'- c:-•
... .:.,
CtlllrOl."~e,TICM
I •·-·... • . . l . ·. • ''° .· -. · ·•,'b"i= II ~-,r=
~
·o ·' 1/\
':,~Ii. D\'1/Q . 2 FO!e AOt:>1TIOt,..IA.L DIM · 'o o,= Qlll!D!!:.!e -~
I . I_.. I I . 1_ .. I ~ ~
C ~05~ 5ECT\Ot-...l ~AL~ : 1/,z_"• \'.O"
loflD 01.A,.!'»~1'_A_c::;_~
,---1 I I I I I
c.111:O•~
"='.NO DIA..Pl-1~....__9_"'"4
,ulba.
--::r· ·«!
." ., N
M\D <e END DIAP~RA.y~ ~A.La: Y2."• 1·-0·
+- Uft
------ . ' - .. .. _ ·- ,_ ·-- ---00·0 ove~ALL U:.~C,.T\..l . I t -~~ be - ------- ---- - - ,--
- rMID ~PAi,..J( 'e>YM . A.e0UT ¢.)
40-0· . -- - - -- ----- -- -
j I
I • ... . . . - . . . . MIO DIAP\..lRAC.M ENO DIAP\..ll'::A.C.M .
J . I 'ir .•
A:,,nts
~ ~q .
'
·7 . • I
I 4 '" - ' (.\J
p; \-\ALF LON~U 01 "-lAL SECTION
~ : Yz"• 1·-0· lz.' )( 11-~ •. ,~z• Cellu l<1r
(°l'Jp. c,a , cor\'\eY- ) q,rd<>r ~• -~ onl\l
I ,,.• 1 ·h0 i _J_ ~optY~ r-4 ~UC"T, TU ...
~•-o• I .i
,...._ D _J
·-t
4o·-o•· \~lC)•
- -- - eo·- o -- C>YE111::ALL LENC.T-U (!OLC.R~IJCt _:,! •i,_:_:'•J
-~~-~~ I&~ o• At(c"'o'2. !)O~T \..PCA.~
-·. 4o' -=o=-· __ _
~~er ~ _-i~ -,,_.,..c~•~J~
0 I ...., -=-1 -- --t-
- . - l ____ 1--_- -_ ----
• ...---t-c-t-+-+-,--i--+-- -.,._-_-_--1-,..:fi .... r_-_-~---+• ----- - • - - - - ..L- - - -
i!~ ! l~ e.- ,Detail Dw9 ~o . 2) _9 _ -i ~· j,17'-o• _J<:o,o f~ C• t d~C r onl,
15 . - - - - - - -- ~ , __:_:_ -__ I __ - ---= =- -· - -- - - -ILi 01· - - - -- ~ - - - -· - -· -~-==----1-- ---1--= -=---~ ~ I_ . I -~ I I - - - -------- - - • ;j • • +--- -- ---- -IL t I -.!, I .
- ~ Loe.a Hon of ~•cm cr11c1< ,.orm«~• '" ' , ■
,"-e,T • .+.--:_ _ --LI rt f"o,,.,t ---- ,~: ,.: - - _'"':X">'G'.\i;;.t• 2 ._,,., ___,_.,.,. - · - - ··- -1. ,-.,:_.·-,_, ~ Leo<ND •o• ococ<-°"." ' c.,~ .... e p LAN . _1<.·l'loncie ~IOCK·out& re~u,red
0 !f·;:flonqe 8 locl<.·0ut1> r,
TOLtRANCI:. fOR t_ BlOCl(.-QUT LOCATIOt,,15 °,- Scale 11
• 4'·0" C:.c;."Tubc:, r-aqu,,wd f o r 011
All locot ions to be meoeured from 4. qirderr,etd tole...,nce :t se• in on4 direction .
Pion obove ~1v•: <D Locat,o~ rouncl an4 ~uora flonqe blocl::.-out& for Te!>t Gorder& IJo!> - 2 ', ! onl4 and . (%) Locot,on:> of 4 St•m Croclc. l"ormer• for Te•t Girder- >,.Jo . 2 onl'4 . ( ,;ho,vn thue, I j . S ee 0..,9 No ~ d.4-for deto ll 6 o/ (g) Crack l"orm•r• and (),) -'dditiono l r 1onqe ~ ,nl'orcement Q<"Ound bloc.I:.· out• .
for Te•t
:J 6
Dade County Transportation Improvement Program C.~ll'~lal Iii I~ I I - .. .,. A c18 •~ , ~o,oprf'ne, 01'"\"1~"'" ♦ t.,O...,,.l"•-' THE KAISER TRANSIT GROUP • joint _,ura
A:>in t-
4'.J?....~ ~-r, I & l'I-IHd'Ai,l • lG•Rot.~ ~ -~ l"L,A.t,JGI!. CVAt, --·-- o, ..-W .,_ UIIIII co-wr, ~. IUCIIUY,_A_ INC. CM!llaant __ ,_, _ _
'fC>I\, ~-ff G, 11Lr>•11. ·~ ----•~L'l'II. - -~-,... - ..... - l«r~ ..,. 6"'",//-,_·. ·:,~
GE.NE.RA L t-JOTt.~
A . MAT,RJAL DATA :
I. Concr-ete : &tra.nqth : f 'c • ~400 pe, i ot 18 Joy a
f'c, • ,4000 pe, 1 ot re tc.o&e
Cement : A STM CtS0 , r4 p e m ( lvl i n . 1 eac:l<.6 / cyd ; Ma• 1\ s,o:..'<--.fa.• Coore.e A99r-e9ote : M 1am1 Oo l, t e - A~TM C "!>~ Ma,._ e,,.,e -a~
fine '-'¥:ire9ate : Miom , Ool, te :::,_:reen ,n9&
\Voter- · Potab le
~ump , 1 1'1- . to!!!>" (Mo><,.,.,..,m) Admixtur-e ·. Daratord .-.C - f o& rT">unufoc.turec.J by
\V R. Grac:,.e Company ( "S.TM C444 - Typ~ D ) or appr-oved e~u ,val ent
'2 . R~inforcinq Steel : ASTM A<..15 Grode <oO unle.,& .::,ther w , -;,e n oted . Golva n , 1.e onl y (1 11 m, 1 t th ,c.:.n~s.s ) ;,t, r ,up& MK. 5~ 0 1
MK. ':,WIA and """- 5!:02
! . F're&tre•:, •n9 Strand : 6. . Low Re lo~oti on ( Stab , 1, 1.ed ) 'i ',; - 110 "- LO -L"X ( l" loroda \Vi re ~ Cable Co ) or qpproved e~u1 vo1 ent
4 . Structural Stee I · P !ate& ond Structaral Shape& ASTM A-~c;,
Tub in~ A5TM A501 ; fi_, • ~G. 1<.& I
.,jot· dip Galvani-z.ed anly ~eor-,n9 Ft. No I o nd
Tub , n9 A&&ernbl y ~ I-Jo 2.
B. TE.ST CY L11-,J DE..RS
5 i,i. Standard Te&t C41inder6 per truck.· load of concrete . Pr&ca•ter ~II re11c>,n '2. c 4l1 nder& p&r true!(. - load to determ,ne re leo&e and '2&-doy e,trenqth and ship rema,nin9 4 c.yl,nder& per tr-ucl<.· lood to test \olo,<l\ S\<.ol<1t: to determ i ne c.on,pre•• ive and ten•ile &trenqth• anct modutu& of e l06t,c. ity . Te&t C.'-jl,ndere. to be ~hopped t o lab :>holl oe. pocl<.e.d c are+u ll '-j ,n ""'ood er-ate .
C . Pl20DUC.TION DATA
Refer note• on o,-.q No e,
O. DUNNAG~ PO l "'-IT~
80°· 0 : .-----, so"-o"
, -----------i
.. f u I~ 1e·-<o-
1 49" i
hl _ U _____ ~ Sto~ , n _, t. ~ - .-.'b 7:a __ ..12
Our ':'9 -j lo4: ( Min ) -j"~ l-lau l 1n9 • l(o ( Ma~ .; «
Tf.ST G I ROE.R * I TE.ST SIROE..R5 2 t, ~
* C ant , le ver D,men .. ,on X. dur, n9 h ou1,n9 can vary from 1'·0" ( min . ) to 8'- o• (mo• .) It ,g de6iro b le t o reduce d1men&1on X. o& much ae, po&eible .
6. F'recoeter &hall prov,de o~d . dQlw«r lo u.,,v«rS<"J?f M1on,, 'oundl~ O'( ,i..-n lenqth~ of st'0nd • each !:> -0 lor,q ,!!oel&c.led at random fron, so poet. u~
FULL- SIZE DOUBLE TEE TEST GIRDERS
PLAt-.J, SECTIONS t, GEt-.JERA L NOTES
-.. A& foJOT~D - fiCl- I a
r lt'•o•
1 Level] ~ Gil!"'CMr .
C :. ... ._, io \J
' j
. I I°/ fll. ~4
1&R
,l!Q • • a~•,· t '- ,'t' -~ v~"-~ -
• 1t • _'t. C:MAl'I ,r~ ~ ... ,'i' -~· I I . I ,. 1011· 1·-,· ~~
I .~ I '.1,_; 4
Lev&I Lev&I
I 0 ~·-c.. t'-c;,• t'-c.. ~·- c..·
I ·111
~ "o ii I -.
-) I
c .9. of s•~•it ,,,,-/4'R Gir-dcr~ I J II I:;..._ - 'r
' ~ :_ 1•,. I' Chamfer G11-•der m
~ ;: ,.◄ &
i'i ri'i 11i i'i 1. 10· j . STRAND E8ITC.&bl AT MID ~PA~ (Strond Fbtterr, w,ome up to 10'-i;' ea side (hold -do,vn lcxat1ons ) o+ ~ Girder ( M,d~n)
-
54-'H. (2101t) Low·Relaw.ation (s1t>biliud) ,tro~ at Pi kipe eo . (D .. iqn Initial l'orce • See. Tobi& D...q. ~ -2 ond f.Jote t-.lo . '2 ~- I-Jo .&)
'\ . r
11 l / /
. .,. 'c,
' I ' ~~~.3~
I ·◄ ta~ •,..., ,t:.'.e".,.•c""-=l•o 'TO ""IJT&
t 14"~La.&1& CO ·◄ ·'ti>< 4\"'~ ... ...,,11,: .... .,. ~ ~-,~M)
I w "IIT ...,,_.,.. (1•-un. ..
:E a.-~--- c.o.-c•m, 110CMaM ' I 'in •u,~ • &,.O ~ :r~ ' • Pi i• r...,_iil"ed de&iqn Initial force per strand ot rnid&pon .,. ~•~• pov1&, co• _ qin::l&f",
.!!,. [ c .q . of Strond• ' - T A&l.f. _ l"OCZ De.$1$M INITl.t.L ?RE.ST R~5SI loJG ' \I..
r ~,~ I
- - Initial 0Hiqri
L3.,.~/, · Specimen Fon:& Pi Rf.M•QK.~ : ,,~ I o.- Girder in Kips,'stranc,1• '
I ~·-- ~Fo&
I I -◄ • 1 ~etoi n in Preco~tcr'!> 'l'ord to • ,,IG,, ." Monitor Comber Gro,vth o,...,d
' ' I 29.'ltc Total ~ ~f"26 ♦ I •~ I Lo.• of - Pree,tre~!»
& in I ...
-◄ DE. TAIL(c.••l'•• ... C)M\.."()'
j ' Send "=> Te&t Lob -- A.q& ~
CURB I 27,'fO k !!>f)'cime.n '-Mimat~d to bC ~o to
~-e. Plan D""~ · No . I ' • 1 40 doY- (t) \/hen 1-•t commenc•!!> I . .,., .......... I S&nd to Te.t Lob. - Aqe. of S\
,; 11 2~:2 1 2~· 10· \ S~c.imen f.stimoted to b& ,:::Ju ~ 2&. 101<. ·.
STRAt-JD PATT .:R~ AT t ROLLE.RS tND HtAOf.RS (5,.,. l,.Jnt~ ~5 O,\tS . !Jo.") 1'Q' 1'0~~•t when te11ot CO<Timcncc.s .
T-..,,.,.,.1>~ Dade County Transportation Improvement Program - .... I FULL-S12E DOUeLE TEE TEST GIRDERS I .f. . lt,,ono . -l/ :!,l;)/"/e ...... ~ - ~C""'"'- ~-~-·------ - THE KAISER TRANIIT Gll0UP ...... _._ :& .... los ~--1-::::~ • ~--, ... ._.,._ • ...,.,cu----
-~~ -~ "' - """"""' ----~--- --- ,._,·IIICIUT,ICIMII.-IIC. CROSS • SECTION t, STRAI-JD PATiERN ;· ---·-- c.------- .... ~~ " . --..., .. .,..M_,.....,..ff, --•-ua ....,. ___
I\ ~+l ,. F'.._C'U,_,,_ __ ,,.IT. -ND--.:.:...11f'
.... ,,.,.., . - .... L¾ -· ..... --- - - ll/~J - S/N/M ... ,,, .• ,·-o· 1- '2 I FiG . . lb -
. 10
2-0A~4 ·4 M11 ~ & • (c&WTC1'ap .,...,.,,. 'OIA!D\4~ .... )
r-- --__ .. __ ...... _-_-_:-,::_J==::!:a~===,~---'""--- ....... •, J
=.:::::::~ J I
~P'-44 TYP. fA . COA.IIIIR 1 ALL CJ•~Of!R:S kddi·honol re1nt ~m«x:iul Midepan
11acL.
T ---------~--1 ,---.---------- 4-ia4-l--;---,..~
i : .. ,. 4-0 I I -I L _______ ---- ------
_J ---------------------- . -- ~-o. (~p.)
=~
HN-C PLI\N - Tt,ST Gl'2Pt,R t '2 ( ADDI rlONAL- R~INF. ~ BL..q:k:.. OUT'5 ♦ a:>RIJ~R~)
~ - 14"• j'-OM
1't-"c1, (Mtn.)
4F2<o~
. " _1
TYP STAGGER OF TOP Bfl~F 5 FOO ~ QOJTOM 4 f IGp
Foµ. ':>PACI IJG see DWG * 4
llTfRNAiE LA'fOUT WII~~ !,AR UIC. l'ALlS OM O'nlE l ~IC>I 01' !!11"08, "OTATI ~Foe e.Y 16()• ~ TMAT T,t,,CK wel.D C.At.l be MAD•
•
I -• i
• I ! i
I
I
• I I I I
• I I I I
• I I I
't' .. •
TYPICAL
& • t. ~~M~M l...\'-lll. "A''
-4f-~ I Hm <w ti" J' I 'r- 4!"4-1
• •
ll. r -· •
tr 2 -&~O,
MIQ -_QIAPl-tRAGM RtlNFQRCING e· l'-o· 10•
11t Bot. d cover _J ' 4,..l<i s~eRao I
(><• ""l""''"'-_,,.,. owo,l 1. 1'·w• 1
1- G.
1 ~
4SI~
48 IIS
1 2"cl.cover-
TYPICAL RE.l~FORCING ~- 112•-1•-o•
I - -~ o- t- , 1 i,,_--~--~--- · _________ j _ __ x --·r ] ill---0 ~-:~ ;;~----~~-~~~--:_---j,: '4-- '!C (typ-)
o • a e,o . (1yp-) t • •
~-J'-_Q'
HALF PLAN · Tb:QT GIRDf..R ~ .3 , ( ..,_DO I T'ION~ R.~INF• FCR \3l...OCK... OLJT'=- 4 C.O~IJe:R'5)
~ -'+• -1°-0•
N.QIL_:
8' Dom ( T4p) I. Tia ~IS (•~•r.-uf>) to "t" f' Ion~- yqtl1·)
foYc,,1<3 \:>Qrs 1n f'~¥1'3• (4i:e.i; • ~ftlS
Z. ~«Inf loqy.5 4f'Sq TH•U ,4.1"4-1 ,.._..._ .A.$>'PITIOl-4AL •L\~!"CMI.C.tN• )=<>• i-e.•'T CSl,_O ..... & .... O"Ly .
t, z~• ~ 13.0 . for Te'3! Gir-der., _t? ii,, f 3 only ( typ-) ~<e 0...,9 . 1 I for l oc.ot i on~ ,n plbn
452'2 Cont ,nuou~
MATC,_. \..11-tL •,:_•
C.F44
- ~ . -~-- - -
71,:--------- -♦--- ·.---=-i---~-~-71 i: 1 ! . I I I
JIL__: __________ j-_____ L __ 0+-_____ _j I --------------- --- . - ---- ---- _l
-4-F4-I
/· t: ,
' ~,
I :e,
~ALF PLAN- -n:.5T GI RDE.R ♦ 3
u C
lU
I),
.s -Ci£ \) Cl
7
(A.t:>DlTIONAL ~Elt-lf, f'01l. ~LOCI(. ou,~ ♦ CoRI-JERS) , '5<:. : 14,"• 1°-0"
IJ',.wu«(S~ .. "1t!f,,t-Jr:-+:--::-::t,-.,..t----t=--.-:--=-::--=r-:-:=-==:-=:::-,cr===I DH.... Mff
I t. 1.brono . ~/3CJ/18 1 Dade County Transportat~on Improvement Program I Fu LL - s1 z E Dou e, LE TEE TE s T GIRDER s - ~-
i!S.,··,
- -THE KAISER TRANSIT GROUP • tD1Rt -'-~-- 0,- ,&. ..... 0Clll9'M'f --•--L'III. - -----•-INC.
CAM - ----IIC. ____ ,_
_. 5 //1/74
MIDSPAN DIAPl-4RAGM OETAILS e, TYPICAL REINFORCEMENT
..... .A, slbww - .3 . I IC
1..Jot e : Mo ... e st ,rrui:- o nd , or tro n e ver&e ~1onqe. re1n;orcernent norn1n oi1y 1n the. ev ~n1 o f Cont 11ct w ith flan9e. b lo ck.. -o~t~
:oc otcd ::>n Dw9 -....o 1) -::,O os to orov ,d c ,::.,ear cover of ~4 oct ,vecn -e t-oi- )' d eayc of olocl( - out
"!l ~-:..::-y- -=: _. , ...- , ,- _1c ~· [ _ ~_:"- t. nd J e '.::i . S h ee! >Jo G, I; t-Jo " l
'5f4 o ic 45 ,_.~ _ S ~r-,A.@ 2G = ;;o·- 4."
1
,z." 10/; ~•' ______ _______ _ __ _ 2'7 5PA . e 10''= 2'2'- ~• ___ ~ 3~4." _ 3,11
'.:>TIE'E'U~ I 12 -sr-,A . se"- e,'-d' ~ ... ) ,;,(;) s1:· e· G c; 34 5PA . <ea•~ '22'-e' _ ~ «. 2'8" J 5f08 _OCA1 i~ - I ?_ ~8"= 8 '-o .~~fbo7' lr" ~~-:+r·"ts 18@, 8 " . 10 '-o" ·-_4• 1'-61 4.~e· ~ 9- ;~-='->'-c' ~ 2 <1 0 ~ ; . ~ _ :;GA,; \1 , I I t"- - ---+-r • --"1 I -- - "1 I ~ • a• ..... - - - - 1¾ ' I '" I I ,I I ' J -,
_'. , ;; ~ 4 M •~•~• 1 ' ' . I ~- / " " /? •}; ;,_ -- •.e0
_o, ,m",•c•'I ,s •oB ,c , ,, :
J - -
.,. : -=,---+..,..+,r- / ----~1------~ ------ ------:~aal=,H+t===+;:,....._ 4.f I'--. - ------ - - "' I¾
--+-- ----- - --- _____ ..,_._ ~-------------- ---- ------,+------ - /
:::::::=:=:= ~ ~ :_-- - ~1 t; B~ia ,- ,4.E 37 1-.
!~-n~a" r - ~
\O~ _t_.i
~·
• I I
I I
I z - "
538-~~tttt~ii~~i==~==~~=~it::-::=-=========::::~~~-~===-?'-~~::::::=,,;;;;... -------- ----- - - -- " '--- - - - -
- - - -=- ~,__--'+~-+----'-- - -----~'--" ---- ===-- ===- -- --- --==---=- =--:--_:::- Jl "'re,
~ - -- --- <l Stro~de Pattern '1 For Hold Down 1
~ ,
( 21 Strands per e,te.m ) Ceto,I& .,e,e D~ O 17 - I~ ( 'ZlO k ; St r Jnde> ;:>er Stern , _ ~ P i K i p s eo (1 n 1t ,0 1)
-·- -
"-.IO'Tt.5 '--- 41!:0,4. (~i!cE. i:>ltl 'tll!T.i.11-) ~ ~ 1 .• A II rcbars .torm ,n 9 re. ,nfor c 1n 9 caqes shal l be ode9L1:J+e 1 y '"1e.d ,- ~ * t~ther wi th t4 1n9 ,vire as per '!:ltonoord spec 1.f 1c.at :on s Toc.K. ~ -= _ \Ve ld 1n9 of only ·starter· rebor~ for coge& \Vrll be permitted ~§.F=-,.==-=#-..==,s;a==..=-=#-,;,= ~UbJect to prror approva l of t he t.n9 1nce.r /
-+ -t- 2-"'S.2' I • l
µ I
h i ~L
I I , . I i,
~1 H / f/
~ -- ~ I - . jN -s::::.s::=s.::::: "'-~
" n
0 - I
- ~~~ I :~ ~,
,______ _ _ IO - Go _ s- Tobie on Owq No. 2 k,r mo:in ,tude of 1-lol d 0 1m ~ r-•Qu1red 1n1 t ,a l d e~1qn +orce. P i 1e.. ,p5 per
I"' ..to ld ()o,vn .. trt:inde, .
♦ '2 . _ ~c t,,Jote 2 D,v9 I to,- 'f)lvan1t.1n9 of the.5-e r eba rs , •. •~- Cutt ,n9 le ,-, 9th"' of rebo,., M._., 45 22 _~~25 Et- 4-F2eo - ♦ 4-e,po-:er b u,·,. · 7 ' ior· -i • ·- "--'-'cldcd
represent re~u r ed 1otol Jen9l h c,-.c lud 1nq lap~ - ~e ~ s .. ,..., , ... p-t { • o r-QCt«. to'f"'rWY ~ .• t ' ·3~;' 1re, .. , ,1 ,,._"'2'Cl. ,-: d .
L,.EGE N D
BAR IJUMetcR7 LOC . OF eAR I IJ GIR D~7
5 1Z.t Of eAR ~
MK'.. No. = 4 E. 14
HALF LONGITUDINAL Scale 1 '2° • 1·-0·
ScCTIO f--1 ·onq1tud 1n a l rebo r~ sholl be mode f rom m u lt iple ~hortcr ~ • --1 '"1-:::d eterr11 1ncd by len9th of r e 1n fbrc1n9 C09CS A.'-.:, ,d op& thru m1de,pon d 1apn r-a9rn Requi red o p ' e nqth • - (., · for ♦ ~ rode. and , ·--o~ f or + 4 re.bar•
t ◄-- ~<lf~f,4-0.,4-li~r q1rder .. No-..24?.ot'l \ 4 -• D""1 e,
. I i<E. 1 t--JFORC \NG SC\-\E_O ULc. (QU~~RT~is1. Pf.R) of re bara una A ee.fore +'obricot ion .- -: ':: :-: - - .J
re in,orcement coqee, , p,eca.,1e, ,;houl d STRAIGHT BARS BE.NT BARS t. t >JD O , A PHR.A GM verify a ll t he bend,nq d 109rorn& and cutt,nq MARK. S IZE. NoRE.Q'D. LE.NGTH" MARK. 5 1-Z.E. LE.r--JGTH" F • F _ ANG E. lenqthe, to en&ure fit , m in imum cover and ._ _ ___ .____ _ _ _.__ :'>E. l l :'>
4~ ?!
M. MD C A P µ R AGM detoo\5 ""' &hown on the drow,nq"' 4e_o:-, -- -4 - - ~7" -
~ . ~ ll 1i C rac. k..
1T1o+ch ~hope. o .. !"'o- mer
, rm
,CRACK FORME..R TtST
f o r Location see P 1 -.Jr
GIROE._& )f' Croc"
'.),vo • I
C E..- ' A\ L... *2 ~orrner5
5 • SH. M r- H ?(\ ~ ·-~ +- _ :, ':'. I -:--_-::· _· . ~ - I - ---~ I -: , -:. -+ _ ------------- ··- ---- · -- ------- _ ··:-,eo I ,. I , I ~-w I AWH\ f==j j ,~ i=t" L"j BENDING DIAGRAM -:-~:~ 4
:\J () 7'. ,) • Lt I· ~ ("o'--:P;,t-1.,.i J all I iii ~ I
T -.--.- □- 1.,, "' ' \ i \,,~=' '1,i :v .. -.!.-!. -~~. ·
, ~ I ~
j - - j i s ~ L 1'4-"(!'f'Lp) Ill ' LI Ill Ill
~r" 2•-3·:_t~ 4El7
""1$1 -\._ i
~;! ~ .' I I ,.._•J,.!~ 3 ' t!-1~ ~E l7 ·+- ... 3E 17 4- 4El7
8'- 1 ~:>'
•~1 c----1 ,_,1 , 6 ... . - ~~~ c.'-IS"
4f"lc.. ♦ 4FIG A
NOT~ , Al l bore, dimcn5ion~ ore
4~~1,o ~e.;~ ~ ~ -~~-~~- 4=~~ ~ - - ~~- ~--~;~• _.________________ ' ------------- -\# 't- ~-~ 4-C. U => 4 '4 U 7--=:,
I~ 8f24 6 i. II-lo' __ ~ • 1'.11" 7" ~ e. ? • a 4 a·- o · w
1 +-- --+ ~
I 4~22. -4- 4- + 72!<.i, !,ts, 14, ~ 2.4 41. ~- ... f -- I f 1
t!J 4~2?> 4 I'- ~ o~d' - -
~ Gs;~ ~ 1~ ~'-0
~· sf.01 • 5 u 9~ 4• :! [lj :r},T : D .3 I I G.-8" ' - to1 ih 3 t3 • 1s•-o• s !:..11 5 1c;. a~.._~ L :!> I<. 4~Go• Sf-18 5 14 J'-.t• - -4 c. • -rs-c' seo1 ♦ ~ e "f-<o'
1,_ 11.. , 1,
1 r: I 1,
:9 4 !4 4--C.:' 51:.OIA ♦ 5 8 "'I'- 5' -t4---'-- '"'t 41---~
,, <()~-2~\3,4- !UUJ
t- -- 7 If) \'I
I r~ ~ -~
(;~ 11 ~~:;,~
\\I 8f\
~ ~e.
c:, GS, 4E?,7 3F2.5At ~ 3,.-~ 4F2C.. 4-f:9
4 ,------ ~ -4 --~ ---~
5 5 ~
5
2.4 I 41.~•
--
u ~
I(;, a~.._ 14 -.t e - <o' 8 .,.,_ 5' •J"' -P•~;;;;, - ... ••- --~- _,
◄ l'1 ' I --- -- -j" 1~3'lt ,~3•~· ~ •~
1
\~'3'~+ --~- I .:i ~ --~I.I Go '-0 I c.,. c..uc... 4E03 4Ml3 ~eta.+ w-.1-. 4
4f-40 4-t 4-F4-1 4
cS ~
~
.. ,4-0 z. 7 0
1!-r.." 4M:',4 4 ~ ,o~o•·
ir--1 r -r 1
- 4_ f'1 :. :: IT 31.c," • (;E.01 ll I I I 4M:',4 4 1 a I Y--8" 1 ~~42 -
I 4--1 4 4
0-rq! -\J]~~ 1[_. . ---]•; i 4M'2\
1 1 1
■ ,1-, 1c;
I• 'I !E>ll 41 ('rff'.) lz:l_ I 1'el- • : m~ I 4, ' I , ",,
o'?,°' 4 4
.1 4
1"4, 1'2 '-0't e, ~·-~ t
2 ,.--: !f
"' "' - 3•
o-l L i----"~ :--:-4:"" I f- ilL
~•-1 •,.J
DIAMETER Of P!~ f'.n; BE';~ P.ARS
~~~ 7-,rrRRUP I e .. ~ ff ~ 31112'' . 2 1~
_ _:: ~ 4 r -?_-., l: - ~-- • __ -
5E_t2_
out to out.
~ 45 15 ~ Su ~o\'C ~ i-- 5ee foJote 3 It (;?,ee i,.Jote '4-
e,M09
Co F44 4'f'I<;. 5F08 ~MoS
8 §_ '-5 Ci,
2 ,,_ <-" 4 4'-0 "
114- 9Ld'
"'- 121-0
'2. ,1. s
Dade County Transportation Improvement Program
SFO& (g F44-
~ 5 ?.'2' ~~ -- ~ C . - · 4 12" - ~e - I c.• -
IJ:.r,,.....,,,,IE.p _lhtf,, ~· . I 1-lriitiv" - .... ~_ I,L , llo,-,no . ~1,...c;a.1.;,._-.,111;~ ..
~ .... p,t,~ i'.'I -~'la~ flP ~
~~~~~~~~~~~~~~~~
4M1?,~4~2I
""!>& THE KAISER TRANSIT GROUP •)Dint..---- -·-·-•--=____ ,.,N:-
FULL-SIZE DOUBLE TEE TEST GIRDERS
REINFORCEMENT DETAILS ti SCHE:DULE5 -" -
r-7/,~~ ....
v ..... 7l\
• l•n~,,.·
t':=%':"'!-:t~"==-=--=-=--=-=--r.-
••n, .&:•lTT "-""'• - °'"'-"•i4 - - ----·---~ --•--LTa ---~--,... - x)~/~ - ~ftl/7' --- A:!I noted 1- 4 I F1q. Id
~ Ill
~
~e.1& ------------- ---- _:.!.If• _ ____ '!!t aepA, •L .. •t~d' '-1::.04 21.!!>" ---- · 3.;A: • 4"• 1Lo• - ~ - -
ell"
5~02~ 1'--G.'t" •••• 4"
~ £ 14- s· i I I I 14',c \0 'l&U'2.1~<. F>At:> I
(To •• ,-u11a. ... ,.~1!'0 ev -'"""'=::;;.q.:~:::.:..;;_==: I _ I Ta•T ui.•,) ' ·
. ., .. IO
~~~~_..,------t --0 l A.. 71 -~ w _:"j w~.L.. :.J I
11 I I 11 1 I I I 11 I I &
1 § ..J
1 • Ill -_, I !!
rT~ -;,t. ~t
ft4~: iifF
I
- 11 , / - ii ii ; I ,, Tl
~ f ~r---: -_r=r____.;_1,Il?':t11'1t'--~.t_a~~:;;::;:::::;:::.~-.==1~-1=== __ ~~~~~~-~:1:::=-ii~l~-;::===r~_{4~t r ~ u ! -. ~ J
11 11 II 11 11 [::=I [ _ _ IL _fL:=JI
, I
~ 11
l u w ~
·o
~ 0 ~
~
;. . a
? " .. . NI•. :c
~ _ ,
f (\J ..
:v '° -· (\I
·•• - - --- ----+- -
=-◄ ~--07_
.... r-
~, ~ -~ -Jt _<?._----+---' I
·o -·
,0 fl
~ 4li
I
L-. . -- . - . ------- -
~
2,i I I
,•i
I I (
J 14- - - --' - --__ =-r:: liil ~:H~~=~ ~-: t-- 6El9 <i.E42
,_, ' . ~ =i:t=ttt ll-:-, .~:~~
r:= '~~ ~::; ci ltl [!l
H
:1:t-t:i:I= la- ½- ----J
~E?l3
--tt l )
: NJ~Tl~~31 _
·•! :J 1 -j .1
•o ! -.!.. . : ...
II : ()
·-.: ~~ . •• a " i()
-llili= = t::
·t II,~ I Ii ·~ :?, ~ -i-
1 "'It -+- _____ ii
,1 I► 1!.11' tf
♦. • 1 ~II! I&
4-0~
a•4•' . ♦••
IO'a" ST-12.UC . 'TU~E.
,~s· · --rJ 1!.o"
~ 0 ., II')
'1, ... 0 -Ill 0 .....
Ill
" • •o': ,0'1 • l • 1!. 3 •1 10 • i ,I I---8 . - -
6Foa -•~2· Cl.
l. 'i' ~~"ct, .:;.F4-4-
I ,- c; \
-6"
1'-10·
4l!l1
~A~c~&
+--- - ---
.... , _t;--1 C. E.0'7
~"-27
l,4x&!-c:.."• I. u.. Lir.o·4A 6 , 4"1t~-•~
- - C!lM~"
1'-10· ~
~ "-- !!'IE12 -<1>E.42 - 4E04- If ,1.E.o~A
,4. ll ?>"7
~l=.O(;
8~28
A.f2c;.
PLAN -END DIAP~RAGM ~CALE.... 3 ''-\Lo"
® 5E.CTt0 N ee"''-• I 1i'=, Lo"
.MC..,.-Js.r,.~ t,r-t-:-:--:::u-:?t--t..=~:;.;-,.-;;;;;--.--.....-,cr:..-----.r7.----i __ , .. .,. kf>H£'2 - •tta7 fl.f>k,_•1:J - r/,,./.;. --s•
Dade County Transportation Improvement Program THE KAISER TRANSIT GROUP •joint_,..,.
FULL-SIZE DOUBLE TEE. TEST ~J_RDEFQ END Dt.TAIL5 - SHE:.ET ~--........ Oil ...w " UIIBt ~
- WE- a ANOCIATU LTI). - .... -p~----
-.-.-•.-.....INC. CAM.-nt __ ,..._IIC. - ----.... .S,//JP/71
......
. ·---- -- - ~· ·
- 5 le
~
?.E \7 ~&.WT C::11t.o~n. • I ... 2. OIJ\..'( • ~ .. ,, ~W::\.\T'
- ,:; \ll'01!~ ._ 'J _ 0111..'f
ce.1...1...u'-"C. ~11.0flaH•
l/z"¢ (270 K) _ ~T ~AN D~(iYP.)
' I'-
LEGEND eA.""- NUMeERi
l..OC - O .F e,,:,,..111: IN Qi~~~lll:
1!>,-.-. 01W t:>..._~ "1 M~ -No.• 4 14-
E,. END DIA-Pl-l~AGM I""' Fl..A.NC:.IR
M • Ml D OIA.PU~M
~,. ~T£M
,l ·'
')
~·
1::- 4::>TEM 2 '-c-" ------- - - --~-! ~T~UCT. TUC>~
'f I
I
J, I
t '.T,
\ ' -'2.''
o~u
/ 0
I o
I .. :
i I- - -,, I j ~E~A7 . ·v V2"9eA
,oa• l - . .j
~ '¥4'
1•-4•
I' - o'o/~·
14" • 10• e,~ARING f>A.O
I
I I
I I
(To be furniehe.d ~ Te&t Lob.)
~· - I ~'Y IV\ . ,A.t)C)UT ~
® 5ECTION e.<:ALI!:.: !>'•l'-o•
I
1
~,coaRt. t.-e.TEM ,.,._.., Jc,,.,oac ..
D
,_, _,
~
·o J)
Ft. DETAIL':> l DWc. .•e.)
i1
I
4E 20 /\ L~4~--, I
\ ee:u ~ I A.FA~
1-------l -~~--l ___ _ ___, 14--'...__, ~
I
- I ~i o ,
!P'~o' • S~-HJ'.= +Pa , ~_'... !
/i Tl l==::;:!:!::::::.j ~, : r~ 4 EO?;
I ,. - l - 4-e. 3 t I
I
' I :,.,
I I I --I
j.l ,0
' N
G»E4'2
....... 11r ANc:14o• ...._._
I
:,-~
-1-1 I i ---l 4E~Co I
4 1 _ ____,_ .4-E ?>I
I ~~01,.. ~eo,
(<:;A.LY.)
(C.A-1..Y,) ' I I , , _). ·o
41!7>,..,...___....___.-,,, -Ilj ' ', -~14BI D I l l j ~ l-4EOS
. I ~I .,
• I -0 I
0 !
1'
~
UO\..DL !Yi
. ~~~ , --
4~04A "°' j1
:; r:7'TH~r~~~-=--_:_---~-~~:_~ ~ ~OTE~
<.e:27 !!~Yi }6!~ ~}t~~aat.a "=>YM. A.t>OUT~---l U-c.,eo 7 <f-~...,_ ___ _,_lo= •- _ _ __ __ _ __'2.'-_\'__ __ ~ <oEO<o(IHI-IIN~)
\. M.& O"WC, . _6_ FO~ ~ OCTA.IL~ -
~ '8ECT\O~ 2 . ~•• DWG -Tt -4 FO~ &T.A-1--lC> PA.TT~tll:W &. ~Ii:.~~ - ~O!le e,.t,..,C ~lt~DI~ 01.._C:.~ -4 -~1!.S. OWG-~ FO~ l!."'10 F\~l-e.M
~LJ!: fl•\~<:/'
..J ff E.. ·3tw;,4 r.::::- -=r :r -:z:: - ·•,-:, ,, .. r u I~~ ,,.~ I - - •-•-•-• -- -•- •---
e County Transportation Improvement Program . FULL- S12E DOUBLE TEE TEST GIRDERS
- -THE KAISER TRANSIT GAOUP .,... __._ ----·-·----·--L,._ -
"°"· .....,,~1.-.IIC. ~----------~---~ END DETAILS SHEET '2
-•~1-loI•C> - G,
•19
-c:.
'"IP ' ~
I 1---l--· I I
4-tO ~ -+---<i.b
2 1~
0
clh
I
I
I!) ..J
·o ·.!.
1'-11 •
1111. II "'
~•~e1J
i 11i 1.LJJ c,!>.- Ll a'4 11 0 "<>
'y,
T \, -·
:..~ -~.:~ __ -::::::._-::::::::._=1___ ---~M-1 1--~--=--;;;....- 11 _:-l :._.. -
a ·-1 I~
,() ~
"'
Ii No-t- f\ l~'•io t• ¼' e¼" t ~i•· ~ ...... - ------· ~~-~'IH- Cc 4E.31
&. Tue,-.:..
IVz" c'C"
al!OI .. I (-'1'•wp)
~
toi!06 I I I I
n
(
-"'
4-1!31_/
"'· -~-.. -· -c-j
:~
, -'
~ i'·•' •~ ~~\d] .I •rn
~
@
. ~ I JI :I •• .. .. ..
3•e'.b'\)!·t11·~1 ,-a: 1'-0 ·
sc:cr10N tSC.AlL: !t' • 11-c:J
'\')
1''
LIMIT~ LIMITS ~ CWD 1'•1.'t"
3'~111." OIA,.MRA&t.t 111.INl'OltCl!.Ml.NT
I 4• _J
~ ~ ,.,-=.,-ti-:.==:::::- 4 I:. 20
•£14
,, 31'25' /
I- 4-iat~
I~ 4.~IG
'IO
-.u--n 1111 1111 Ill J....4---l--5t\2
Pl ~l~t11w----&~'ffi1k1111 ,~ 'ic~, ,, :~ .. ~1~ ·
b. t - -')
..,; e .._..._. I ~ I I== 4C37"\
li&>1 --
11iC..L.J)
_j u ~
9
·a: P'..i
l_t-' l-' .. - ~ ~ff-P-7
~•o •~1- 71\_-k,;I-l. __ J &llG ti1-10.1 -~ " O~A~!,-Q"L.G '
(TO . .. .. 0 -- + >< ,~ ~ 1~• · I &t ~Ult \
9Y Tt.r 1-l~HtD I . L.A&.) • o/ --,-..:::5~-+-~'t •_J~,1
1
' - ~1 311
'
1'-~, ""' . ~~·-
I' ;J+I_;, ~ C.
<ti'
,\•1:1 I I <o'!;
.
l=-----i I l-=-------111==:f I b
/ I j Ill
. \S I I
> r _;::;, -:,
-6 2
-e!
2
=5 t t t-- 4++-- 111 ' Ill III ·= ~·~ ANC~ORS
L
IIJ i 11141114'!1l 4'lll.LJ.tE?17
-. , . - -·· .. ··· --
4Eo4 'f' 4E04.A
NO,t:p : 1. 5ee D,,JG, . e ~ 1k. oc,,,..,1...~ .
ill ~l.02 ('-"L-V.) ~ tj z. ~e r;:l!,/G~ 'Z 4, 4 ,..~ &T"R,'\ND A'l.r,~RN ~- ~ e ~ - 4 !"OR &°'.R ~OIN:i ~M , . see D,iG. e l"OR '-ND l"INISH .
LeG~ND
i,.OC.,'\l"ION OI' &,R IN GllltCCJl7 ~IZ~ Qr e.-'\R~ ~'l!,Nc ~
MK No. • ~t:: 14 /
e. • eND OIN='HQNiM ,. - ~L.AN<Ml M• MIDPl~M & • a-reM
~:i~~;:~2
1111 \ <oEO<o . \
<o!:Ol
"~• • J t::=:1 t --b-=i ~ a.'27. ~ ~'Z&
• I
I nJ,1·
- ,. 0 -~ .....
e~o• r
at\ s~c,:19.N
X.7...,,/s'- -a/7, · ~--+-.....,,.Jf...-,.i--+---,-=--,,....,,.-~-=-~-=-=-=-=....,,....--f - MTI
Dade County Transportation Improvement Program FULL-S12E DOUe>LE TEE TEST GIRDERS C.-~!!lll!O $•"le µ::a;ul!,.!.!~~.IU-~+:::=:..:.:::.::....::1-.:=~'.!....::::Z..:::...:.::....:t..a=~~
~ .... ---- -- ·-
-· ....
THE KAISER TRANSIT GROUP • )Dint venture --· ..... o,...., J.. MIIIII CJCllalMT
- WE- I AS80CIATU LTI>
·-POIT, IUCIIUY, -•,-.NC. C.- .-TH - AUOatoTIS,INC.. ~-~ AUOatoTU
.... ...5'//4'/7d'
END DETAILS - SHEET 3
ICM.I AS 1,JOTED - 1
r
~+,Hl tvb,n<J •pacer '1,C:: lont ·
( l"o.d. > ••1~ t.d .) .. cut bolt le, ·fro,., race of nvt
"iv L_j_ _ --~
1 - .c
·-,1 '.; ~ ",:) ... 0 ""'\) ~ :I C •c-t
.- l: r-111
1/) .; 11 D Ii) -- ~-
Cl
!- 1(),---
~c,! tubtn11 5"11~r 151~ ., .. .,. , ( 1'0.d ., ••.~ l ,d.)
J°f' &td . wt pqoe roller ,. ''i tonc:i ,
(1.~1s''o.P., '•¥• ~ 34- ♦ .~pit 11.7~ loi,~ w/1,;_ 'tln-ead (A-~ .sAE.- e or ~1!:.-i) ?..-.'e -,td waV\,r~ • .~d - l'lc,c. nl.lT-
P ,S .G . 1!>·-25, '1/4" 1-\ctavy he>< r, u~ or Of"t-l"ovcd e9ue1I
~O«.-
ID(T. lb_ ~,{_><7"-,,.&" <S-ltW~♦ I ~ • ,.-- !f{K if',< e• £ ._EQ.e fo'2. ♦ 2,
"1MT.-it ■,~,.~$' c.1•0t2 ·• ♦ 11,r.'IQ--.lx.&•a'' .._,_.-.2. a W/ I· tCs/<f t+OL.■ tr'ff)
C::.1:.!.liiiA.SR.. (,-'f p,}
.T~p -
Weld ends (e-zc flort)
JI ".;ti tl C,
. I c< I IJ" I\. I " · --- ·- 1·.- ~ .c,·
,ta'"' i'"' ~• &11t-0&1l. .. I . ~~"\11'"'-c." c-1coa1t. •'2'f't> ~ . · S t' I- I i thick plotic
1•/'1.' 7~;.· 11/4°
10'
S wosher (tyr-')
~~/4'
'O, 5 . G, 1!)-12, -s;,,.• f' (ult. ~~ooo poun-i1~' _ co11hnuou.$ coil thYcc:ided rod or Ofprova:d equal
HOLD - DOWN DETAIL ~cole : c,' ■ 1!if
*NOTE , Refer- to no+~ no . Ii for dco+en•ion ing / &trippinq procedure . ~ .. leooe hold • downs in two opero+ion&: ; ir•+ two hold · down~ on one •id .. ur mid• ,poN i'n one si.cnultoneoua operation ond then +he r .. moininq two hold · downs &o o• to eliminofe twi•finq ot' the A.irder durinq reloo•~, J.l&o, both coil rod• of df,1 one hold · down •hould be remov&d •• ne¥rl1 eimul'toneousl~ o" pouible.
t !nd heoder )( .o·
to!c,' 1o!c,• Abutment -
t Hold-down-+~ - - -_._ ___ ~
l !nd header~
G.6 . of •frond•-----,
r ..,
I. The three +e•+~·,-der• 4.holl be co•+ ind ividuolly. i . See "TA&I.E' on d . no . Z t'or re~uired de•i;l'I
initial ,pull Pi at m 'l:l - &pon of unit. Teneionin pull of obufment& •hall be inc:reo•ed to occoun for fr iction Ion thru hold · downs and heode.rs ·! however, the muimum pull a+ abutment& ehol no+ , e .. ceed 32. s kip• Fol- on\,/ ~+...and . ( ~-!>It~ ov«,.,.,itJ
~- Fr iction lone• thru holdL down and heoder• •holt be determined 1,1 te~+• c>n two repri•entotive e+ronde Ct(?p row and second row) prior to .+reo, • inq_ of fir•+' unit . Thia doto •holl bejn+•rpol•+ed · to 'determ ine the re4uired jock in' force. for ooch •frond . ·
H
l Unit sofM oF tee
..: l'l\' ....... _. (
4. stre•~in : ot' three +e•+ uni+• shall b. done. from ' o e.ricf qnly·. · . 6 r•a• ' .tForios pro~nueiv•l-t ' from top to bottom with c:en+er- •fr'on4' of eoc:h 'horizonto'I row bein~ •froHed f;,...+ _ Mov."'H' • 41.0 kif& t per •+e.m pe.r- hold ·down for Unit .... I. Mo .. . 'T" • •~o kiP,e per abutment . ,.,.. •+tm forUnit N• I OP.pli•l:I 14.1&' obovcs. bottom of •+e.m .
5. Drflpc. of 1'f'rond• u ~•hown on dr-owinqs onc:I P.Ot+•rn ot t roller OT •nd h~ocl.r ... ulh•t tho+ ldimon•ion •~• b.,twe.en end crl' unit and l of haoc:ler • '1
1. For •i,nific:ont voriotion• in climoneion '')I.,'
p R O D U C T 0 N
;'
1 1~4•
:5t~ffc11v ~ ~ '*>< I 516 "-., ◄ Cr,',p,))
(4 pey hold down.
> , .__t,• .. ,· a•~,.oac1,1 '!!j j •~t" !I .-<p -q,- --t-' ' ~~'\c~c.•c,eo-• '., 111a·~ .•.a.• ~ ~ t o? ~,, f bolt -1' ~ ~;,• f coil •,.•
PLAN HOLD DOWN
Seo le , !'• 1'.o'
e
fh r eoded rod .,
1
I ' ~ i 1 ~ ...,.._ •• ,. .,,..__. I -l■<T,.!.(>'&OC-~I(\ • 11 I ' I&.. 7 . • _, -"" ,t:13'1. .,.,.,11.,.., -· ~ l"J•* ~.,._, ,_........ N W/l•ll C't~O... -
~ OJ
"T'JP
ft l/7.0 ~ DeA,. ~·-o· lon3 I
1\?.3/4_',. ,<.,",.,s•J ' • (6AL.v~1zs.-v) ~ C.><3-4- \ong ----t-'_ ~ -
(oo t<cr,- -LVAt<ll:W . 1
..... c .. -• ""'P TofO ~A LeLE V AT 10N_J 1-----~v•'f""-c• oF 101-A"T&.,) ~ ·
PLAN
~I IO I,.. -,1. I I R•S/e'min. 1'. ?,'
~,i i-------+--- 1s/11;· .. 41;,· slot
t V1.', !",. s111.·
"fir,' ♦ Threoded top hole for quide rods ('2 per itr
';i_ ,.... t
SECTION
st · '
~ ~
....
.<-' ::::--d)
·~ '·~r "!.
C,!>/4'
~ - ~ !,/4' ,,,;,£ !41 r' - ft 5/4', ~S/4\ iC,'
I! 1/ (c,:;4l.VA .. I eav.'
r---+----1---.L...;., r- 1's ~t, c.Jo~,,1,111'•· ><1oo'k•~2.~ t-
~ ( - ............. %IE:t)) J
- 4 L'4'4> Ho le
\I I' "ff1'tt -1nsedi:m 1
' R. no S , ......_..,.4,. ,·u;;o)
LE L EV .A.T ION_j
~ -- I/?."~ D~A ~ .,,0·1001
PLA.N
1'-4'
Soc _boc •fo'"''.\->-'°~ ',_A N • =
I
6"::)SECTIOI-J
PLAN @
1!.4'
A 2 ♦ 5-'< 210•,11.1ded to f.!ld"lt In a•♦ w 1/t'dttll' I-Iott
ort• i&e"<f>w.'21.d'~ . .._. i:;~;. · , ~ s E RT I ON re. NC> s ~V.t.HI-ZU>)
~cole : c,', 1'.o" ( li?l!Q'O : 4 ~R GIRDl!R)
(8 ln .. ,tion It.'• 1.10.s for T .. t Girder• "-lo• - 1 i, $ &hal l lo& .t,ipped IOOH 1o te•t loloon,1or-.)
9diu•+ •fr•nd;.oH~rn o+ header• +o ~inh:a in tk'G 1om6 dro e of e+rond•.
G. oc+•n•ion/bur top I& &frond• ('I P."r •+em) bdore ,..leosinq hold· down•~ c:>eten•19nf burn remo,n,nq .+,-.nd• "bf+•r rcol•uinq hold· down . Oe+en•oon irf.q/ burninq eholl be done '°mul+aneou•l'I ot both and" haOd,~•; oleo\ proqreu ~1/mm\'ITricblll/ obou+ t double. +e6 ono frofn toP. td bottom . El:cen+...,c ,+v of P,f'Hfr\'IH obout t doJblc +ea •hall no+ e~ceed I tho't d\le to thre• ~r•noil• (one row) o+ onf 1m~ .
1. Sc:reed fin,eh +oi o +'t•nqH os •mooth ... Hible . All other r-t1on flair. •• and •tc.ms • o ll hove • •mooth PJor m in1,h . .a.Ii' end• of' •II unit& cflon9ee
• D A T A
i-- -·--7 <L I . ,.... . I I
~; 1 : ~ - ;i
E.U'.VATION Li -~ I, . . . P L AT L No I (A5 Sl-40\Vt-4) -t- E.Lt:.VATION
PLATE Na I A(OPP. I-IAND) p LAT E. N ° 2 ( A!> ~140-..,M)
P LATE N? 2AC010.-. ~•ww !>cole , ~••1!·0• (-suQ'o : 2 o..- EAC~ P!!.R G;IRt>ER)
arid •+em•) •holl olao be f inished •mooth . In qener-ol, of'ter deten• ioninq, rece•• a+...and• 111 cow.r- '1'ece&& with ef!O"''t ind th!tn finish •+em• ond t longea.
&. ~pra~'top, of flanqc. with li'lu ioil membronc f'orminq cbm u!)ll confor-'fh in,a to A5TM C~O'I., T'lpe JI, . keep +oP.. flanqe covered' until rele•se of' m~mber.
'I. l'e"t'ar to d~q. no. I for locofion• ~ h•n•l in9 lift P.Oin+•, dun~qe point• ¥or 4.tor-a~ in ~ord ond ~unno11c. poinfl dlu-inq tr-on•+'er r•ilfhuck. Liftin!,I ambe4' it4ma (numbe.r>; diamefer- •r. •nd len.qth ol' •+rond1_) •holl be opprovo-d bl the c"'in,c.r bc.f6re produc+,on comme.1"!:e•.
5co le , ~• • 1'.o·
(R.C.Q'o , 2 ol< 1!!.,\C.14 Ps• c.1aoe.e.)
Uni+ Pr•win411 _ 6calc : Vt"• 11.o•
10. (.INC ... slloll IDt at, •• ., ,. dCl11• old wllctn th•~ era loc4t4 Of'I ro11 Qar• for • .,,,..., • .,t- to tc•'l-la \Do.-ct-or" at- llk•1<1a, h1n,01• .
NOT~
-Dade County Transportation Improvement Program
-TtE KAISER TRANSIT GROUP •J■IIII-'-----·-,A .... ..., --•~L19. -
-. IIIC■UY, ICIIUlla.-,ac:. c.- ..... -~--......... -MIDCMT■I
FULL- 512E DOUBLE TEE TEST GIRDERS 1-AISCELLANEOUS
PRODUCTION
_. .A.5 WOTec:> - 8
DETAILS DATA
AND
-
Fig. 2a Form for Manufacture of Girders
Fig. 2b Manufac.ture of Girders
Fig. 3 Transportation of Girders by Railroad
Fig 4 Hauling Girders Into Laboratory
-~ 0. -~ w ~ en
w ~ w 0:: C..) z 0 C..)
0 0 0 CX)
0 0 0 V
0 0.001
GIRDER 2
AVG. f . ~ = 7950 p.s.i.
Eeode = 57,000 ~ = 5080 k.s.i .
Ee. I = 3850 k.s.i.
Ee. 5 = 3410 k.s.i.
0.002
CONCRETE STRAIN ( in./in.)
Ee. I I Eeode = 0 .76
Ec_5/Ecode = 0 .67
0.003
Fig. 5. Stress-Strain Curve of Florida-Oolite Concrete
-en
.JI:: -(.)
w . en :::::> ..J :::::> 0 0 ::E
(.)
rC/) <t ..J w
0 0 (0 r<)
0 0 N r<)
0
I I I I I I I I I
3850 k.s.i.
I ----1 --------1 ---7-- I
I I I I I I
f 'c = 7950 p.s.l.
3410 k.s.i. -- ----
I I I I
~0.1 IC ~0.51C I I I I I I
: ! l I I I 8 I I 6 81 I I 4
oo I 2 · ) N STRESS { k.s.1.
Fig . 6. Secant Modulus of Elasticity As Function of Stress Levels.
-·-"'
.le -Cl) Cl) UJ a:: ._ Cl)
0 0---------------r---------..----------------;o
2~ I I I
0 0
0
2)@➔
20
EXISTENCE OF LARGE ULTIMATE
STRAIN IS UNCERTAIN
WIRE BROKEN AT JAW
CD FIRST WIRE BROKEN
® SECOND WIRE BROKEN
@ THIRD- FINAL FAILURE
J/2 In. DIAMETER LOW-LAX STRANDS
E5 = 28,900 k.s .l. ( AVG. OF SIX)
f pu = 278 k.s.i. ( AVG. OF SIX )
40 60 80
STRAIN ( 10-3 in./ in. )
Fig. 7. Stress-Strain Curve of Prestress Strands
100
-·-u; ..ir::: -en en w a:: ..,_ en
Or------------,,-------------,,-----------.-----------.......-------OJ
0. (!)
0 ._,..
0 C\l
0
•
5
:tt: 4 REBAR
LEGEND
• +
•
SPECIMEN# I
SPECIMEN4t:2
SPECIMEN#3
■ SPECIMEN~4
MODULUS OF ELASTICITY = 26;700 k.s.i.
FAILURE STRESS = 102 k.s.i.
YIELD STRESS = 61 k.s.i.
10 15 20
STRAIN { 10-3 in./ in . )
Fig. 8. Stress-Strain Curve of Mild Steel Reinforcement.
Fig. 9 General View of Initial Static Test - Girder 2
I J I I I
I I I I I I I I f I I I !-o-? ~
'\ / Tie Rods Nominally Stressed to Floor .
Fig. 10 Hold-Down at End Supports
9"
I ½7-j',; >
~13'-J" )I ◄
A (> B C D E
13'-t"-)Ml•-c- 13'-i" ) Ii t3~ 111
ao'-o''--------------i~ a. Longitudinal Section
I N
~10·-6·~L10·-6·~ _l
Ir:) R 31- 611
--=~L --'--:-- --,,..-? ----------6" T 6" T + ,t_Jfl' , t . Pi I 1
T-1~1_1 ~ T ~ ·y ~ 31-6
11
..___________ l 3•: ~' I '3' ..... , 3~ T r'" 'i:,i r 1 ·b. Plan - Girder 2
Fig . 11 Loading Locations - Girde r 2
I ~ I I
stricity
Actuator---•
Stressed to laboratory floor
Fig. 12 Fatigue Test Set-Up
Fig. 13 General View of Fatigue Test - Girder 2
Fig. 14 General View of Derailment Test - Girder 2
A B C F
---13'-1" >I- 13'-1'' >1 4 13~1"+ 13~""+ 13'-1"-+l 9"
....------------- ad-o"-------------a. Longitudinal Section
l..-1· - 211 , l 3· l I
_J_
~ ~5
+Pi -~ P5 + 0
3~611
..-5
--L 31-6
11
T .N b. Plan ~ Fatigue Tests
10' -1 21' I I ◄ 1a'-tt61
t> 8-I - I ..-4
111
·~ . ~1 ~a--r s"fo- 6
11
~ -+tef+r- _1,ef I I
~ ~ - T 3 T~ I 3
1
- e" r3•+3, !4-34"34 T
c. Plan, Derailment Tests
Fig. 15 Loading Locations - Girder 3
3 1- 2 II 2'- 6 II 2'- 6 11 3 1-211
a Vertical Translation
Longitudinal
b. Longitudinal Translation at End Diaphragm
Fig. 16 Locations of Translation Measurements
SOUTH STEM
Rigid Frame
NORTH STEM
Fig. 17 Measurement of Deformation of Cross Section
~1,,1..-i ~® ~f~~1---------"IL~ 0 l/3 l/2 L
------------------------,---------,N ~w,-=-=--=---=--.. 2 ---=---=-=-~-=----=--=--~5-=========-=----=== =·""' -Ek l ~------------------JL __________________ u ------------------------------ - - - - ---- --
( a ) Girder 2
@
0 L
N ---------------------------------------i----- @2 -----@3 ----r4 -----~-- e5 ----- €~
u_ __________ • - . _______ IL ___________ . ________ .JJ --------------------------------------- - - I (b) Girde r 3
Fig . 18 Locations of Rotation Meas urements
-w _, C\J
SOUTH (LEFT)
NORTH (RIGHT)
t 7'-o"
t t
41-1111 9" f f
s'-011
t 101-011
t
II
+i ~ + + + + + + · 2d' 2d' 2011 3011 30 11 3011 3011
LOCATION I 2 NUMBER 3 4 5 6 7 8 9
SOUTH STEM 18 GAGES
NORTH STEM 9 GAGES
END DIAPHRAGM 4 GAGES
TOTAL 31 GAGES
2011
GAGE FACTOR = 2.09
•
Fig. 19. SR4 Gage Locations in Girder 3
0 0 rt)
0 0 (\I --LL.
0 - 0 0
w a:: ::)
0 t- 0 <( 0 a::
" I ' I\_. I/
w Q. ~
0 w t- 0)
~
0 0 CX)
0 0 ,.._
--~
2
", \
TEST GIRDER NO. I
TEMPERATURE CORRELATION BETWEEN CONCRETE a AMBIENT
+-+-CONCRETE TEMPERATURE NEAR BOTTOM OF STEMS
-•--AMBIENT TEMPERATURE BETWEEN STEMS
\__ __ . -=::::--.;;;::;::~ --....:--.-__
3 4 5 6 7
TIME IN DAYS
Fig. 20. Temperature Change of Concrete at Early Age
'.'i z . 21a Crack Pattern for Initial Static Test of Girder 2 (Center Line to 20 Ft. :;..:ast)
Fig. 21 b Crack Pattern for Initial Static Test of Girder 2 (Center Line to 20 Ft. West)
INTENDED TEST PROGRAM CLL + I
100%
I lI .,, }
(e=5") ( 5") I I I I I I I I
I 2 3 4
:n .nz: (-1.3) ( - 1.5)
100% ---
ACTUAL TEST PROGRAM
CLL+ I
100%
t lt
( 5 ") (e=5") I I I I I I I I
I 2 3 4
m. nz: ( -1.3") ( - 1.5 ")
100% --
Fig. 22 Fatigue Test Program - Girder 2
n I
( 5") ( 5") I I
5
-II: I
(5~ ( 5 ")
5
--(-1.5")
-
6
CYCLES (MILLION)
6
CYCLES (MILLION)
Fig. 22A. Cracking Pattern prior to Fatigue Test - Girder 3.
-
~ 0
IO {\J
.... z ~ W 0 u {\J
CALCULATED FINAL LOSS 23 %
0:: w a.
~ 0 U) ,0 --·-· • .--:3 - .~,::::-:::-:=,------------------ --------=---+--0:: •• , ........
~ ~ r, 0:: 0 ♦,
a.
u.. 0
~ en IO en 0 ...J
30 RELEASE
TEST LOSS
GIRDER NO. I OF PRESTRESS
SECTIONS MEASURED 3'-6u FROM MIDSPAN
-+-+-AVG. DEMEC GAUGE MEASUREMENTS
_..,._.,_AVG. WHITTEMORE GAUGE MEASUREMENTS
60 90 120 150 180
TIME IN DAYS
Fig. 23. Loss of Prestress - Girder I
it an (\J
-.... z it w (.) 0 a::: (\J
w ~ -en it en an w a::: .... en w ~ a::: 0 ~ 0
LL 0
en it en an 0 ...J
CALCULATED Fl NAL LOSS 23 %
-----------·------+---------...... -------- - -~_:;.-- - .. + --A~· __________ .:._~= •-
.,.</A +-
>;?'
30 RELEASE
TEST LOSS
GIRDER NO. I OF PRESTRESS
SECTIONS MEASURED 3'- 6" FROM MIDSPAN
-+-+-AVG. OF STRAIN GAUGE MEASUREMENTS
-•-•-CALCULATED PRESTRESS LOSS
60 90 120 150 180
TIME IN DAYS
Fig. 24. Comparison of Experimental and Theoretical Prestress Losses - Girder 1
-:c u
q rt)
&q N
q N
z lC) - . -a:: w 0 CD -= ~ <t (.)
~ 0
---- ·--+ _. __ ............. ---------
+ ......-:: ~=------------------- - -. _p-•-·- . . - -- -e-
//'T •-
30 .____-RELEASE
60
Fig. 25. Camber Growth - Girder 1
TEST GIRDER NO. I
CAMBER HISTORY
-•-•-CALCULATED DESIGN CAMBER
-+-~-MEASURED ACTUAL CAMBER
90 120 150 180
TIME IN DAYS
0 <l'. 0 _J
y: u :::> a:: I-
0 ------..------..------..------..-------..-. 0
0 CD
0 (!)
0 ¢
0 C\.I
0
THEORY
FOR E = 0 .76 (57,000 /fi;) I TRANSFORMED
CLL +I
UNCRACKED SECTION
0.2 0.4 0 .6 0 .8
MIDSPAN DEFLECTION ( in.)
Fig. 26. Loads vs. Midspan Deflections (Uncracked)
- Girder 2, Initial Static Test.
1.0
........ en a. ~ -C) <( 0 _J
~ u :::> 0:: I-
or--------.-----"""T"""----....----------------0
0 co
0 CD
0
CLL + I
0.2 0.4
CRACKED
CRACKED SECTION
CRACK LENGTH 11-6
11
MAX. CRACK WIDTH 0.00611
0.6 0.8 1.0
MIDSPAN DEFLECTION (in.)
Fig. 27. Load vs. Midspan Deflections (Cracked)
- Girder 2, Initial Static Test.
ct_ SUPPORT ct_ GIRDER <t SUPPORT
12.0 k 11.6 k
23.6k 0.2 23.3k
35.Sk
EXPERIMENTAL - - - THEORETICAL
0.8
UNCRACKED SECTION CRACKED SECTION
DEFLECTION (in.)
Fig. 28. Deflections Along Girder Axis - Girder 2, Initial Static Test.
-+--I a. ~ -w ::::, C, a::: 0 .,_ ~ (.) ::::, a::: .,_
0
0 0
0 CX)
0 <.O
0 ,:;t
0 C\I
UN CRACKED
NOSING a 60m .. h. WIND
0.02 0 .04
MIDSPAN
SECTION
TEST
MEASURED BY ROTATION GAGES
0.06 0 .08
751-211
ROTATION {degree)
Fig. 29. Torque vs. Midspan Rotation (Uncracked)
- Girder 2, Initial Static Test.
0.10
-+--I a.
.:,t!
LLJ => 0 a: 0 I-
:x:: u => a: I-
0 0
0 CD
0 <D
0 ¢
0 C\J
0
CRACKED SECTION
NOSING 8 60m.p.h. WIND
MEASURED BY ROTATION GAGES
0.02 0.04 0.06 0.08
MIDSPAN ROTATION (degree)
~ig. 30. Torque vs. Midspan Rotation (Cracked)
- Girder 2, Initial Static Test.
0.10
Cf_ SUPPORT 11 1 -511
- - - THEORY
TEST
'-1"
( MEASURED BY ROTATION GAGES)
UNCRACKED SECTION
_3_'-1"
13 .7 k.-ft.
_29.1 k.-ft. -.....:
,~91_9k.-ft.
' ' .........
(i_ GIRDER 13
1-1
11
l6.4
ROTATION ( degree )
131-I"
(i_ SUPPORT u'-5"
CRACKED SECTION
Fig . 31 . Rotations Along Girder Axis - Girder 2, Initial Static Test.
tJ
or---------/ A-/- 2.8 x CL L
-"' 0. .:ii:. .........
0 <{ 0 _J
00
w 0 _J ~ X <{
F.S . CRACK
2 x CLL
CLL
CL L = CRUSH LIVE LOAD
~ = 0.091; FrRST CYCLE
1),, = 0.111\ SECOND CYCLE
2 . 4 6
MIDSPAN DEFLECTION (CENTER-LINE), inches
Fig. 32a. Load vs. Midspan Deflection (Center Line) - Girder 2, Dllrailment Test.
0,-----------------------------------....-----------------l()
4 .5 x C L.L
~, ~ 3.8 x CLL 0
81 1 / "--MIDSPAN CRACK WIDEN RAPIDLY -en
0.
.:,,:; I l -·"J 2 .8 x CLL ........
0 I I f.J--F.S. CRACK <I: 0 _J - --
2 .0 x CLL _ _,_, w _J
X 0 <I: I()
CLL CL L = CRUSH LIVE LOAD
0 10 20 30
MIDSPAN DEFLECTION (CENTER-LINE) , Inches
Fig. 32b . Load vs. Midspan Deflection (Center Line) - Girder 2, Derailment Test.
t L/6 t L/6
t L/6
58.9k { ~ 2 x CLL)
ELASTIC THEORY (UNCRACKED)
t
1.0+5.0
2.0+10.0
L/6
t L/6
t L/6
i
80.9k {2.8 x CLL)
TEST I CLL = CRUSH LIVE LOAD
DEFLECTION { inches)
Fig. 33. Deflection Along Center-Line Axis - Girder 2, Derailment Test.
ol- 2.8 x C LL ~/ co
F. S. CRACK
-"' I 2 x CLL 0.
.ll: -C < g w ...J 0 X ~ <
I CLL
CL L = CRUSH LIVE LOAD
A = 0.1211
, FIRST CYCLE
fl = 0.1511
, SECOND CYCLE
2 4 6
MIDSPAN DEFLECTION ( HEAVILY-LOADED STEM), inches Fig. 34. Load vs. Midspan Deflection (Heavily-Loaded Stem) - Girder 2, Derailment Test.
-(/)
0. -~ -0 <t 0 .....J
LLJ .....J X <t
Or-----------------~---------------..-----------------1{)
4 .5 x CLL
3 .8 x CLL
g~ / "- MIDSPAN CRACK
I
I 0 If)
0
WIDEN RAPIDLY --W.S . CRACK
M 2 .8 x C LL
1 I 2 .0 x CLL
CLL
CL L = CRUSH LIVE LOAD
10 20
Ml DSPAN DEFLECTION ( HEAVILY-LOADED STEM ) , inches
Fig. 35. Load ys, Midspan Deflection (Heavily-Loaded Stem) - Girder 2, Derailment Test.
30
r L/6 t L/6 t L/6 f GIRDER
L/6 t L/6
80.9k (2.8 x CLL)
1.O+5.O 58.9k (~ 2 x CLL)
t L/6 1
CLL = CRUSH LIVE LOAD
DEFLECTION (Inches)
Fig. 36. Deflection Along Heavily-Loaded Stem - Girder 2, Derailment Test.
8-------------------
-+-Q) Q) -I a. ~ -w => 0 a:: 0 .....
r()
0 0 C\l
w _J X 0 <{ 0
2.8 x CLL
2 x CLL
CLL
CL L = CRUSH LIVE LOAD
~ = 0.045 DEG. FIRST CYCLE
~ = 0.047 DEG. SECOND CYCLE
0.5 1.0 1.5
MIDSPAN ROTATION ( degree )
Fig. 37. Torque vs. Midspan Rotation - Girder 2, Derailment Test.
O' • . - ~t ;;1 4.5 X -Q) Q) -I Q. I ,-:1 3.8 x CLL ·-->I:. -w ::J
F.S. cf~K l/4xCLL "- MIDSPAN CRACK 0 a: WIDEN RAPIDLY 0 I-
w ...J 0 X 21 /J I 2xCLL c::x:
CL L = CRUSH LIVE LOAD CLL
2 4 6
MIDSPAN ROTATION ( degree )
Fig. 38. Torque vs. Midspan Rotation - Girder 2, Derailment Test.
t 11 1-5 11
t 131-I" t 131-1 11
t <t_ GIRDER
13 1-1 11
t 13'-I"
t 11 1-5 11
......._ --- ---- ----
58.9k(~ 2 x CLL)
- - - - ELASTIC THEORY
---TEST, (MEASURED BY ROTATION GAGES)
0.4+2.0
110 k (3.8 CLL)
CL L = CRUSH LIVE LOAD
ROTATION (degree)
Fig. 39. Torsional Rotation Along Girder 2, Derailment Test.
t
0
o• C\I -,
I 0 0
-fl) a.
~~ .x -r<)
a..
0 <{ 0 _J 0
w _J X <{
1-
(0
z 0 w ~ ~ _J
<{ a:: w 0
0 C\I
0
0 0
0
I/ B 0
El
FLEXURAL THEORY
0
/ 0
0 SOUTH STEM ( OR HEAVILY-LOAD STEM)
EACH POINT IS AN AVG. OF 3 GAGES
0 NORTH STEM
EACH POINT lS AN AVG. OF 6 GAGES
Fig. 40. Load vs. Strand Strains - Girder 2, Derailment Test
2 3 4 5 6 7
STRAIN (I o-3 in./ in.)
0
8 9
Fie 41 Cracl: Pattem of Girder 2 Afte::- Jerailuent Test
Fig. 42a Development of Cracks at Midspan
- Girder 2, Derailment Test
Fig. 42b Development of Cracks at Midspan
- Girder 2 Derailment Test
Jct.
I
CRACK
Fig. 43. Mechanism Creating Horizontal Cracks at Midspan.
i i f i
Fig. 44 Diagonal Cracking of End Diaphragm
Girder 2, Derailment Test
Fig, 45 Cracking of Stem at End Diaphragm
Girder 2, Derailment Test
-en 0.
::x -0 <( 0 _J
w _J
X <(
0 co
0 v
g2_lll ~61
ELASTIC THEORY (UNCRACKED)
CLL
2.8 x CLL
2.0 x CLL
0 FIRST CYCLE
8 SECOND CYCLE
CLL = CRUSH LIVE LOAD
6 I = 0.070 11, IMMEDIATELY AFTER UNLOADING
6 2 = 0.04911, AFTER 1.3 HOUR
1.0 2.0
MIDSPAN DEFLECTION (in.)
3.0
Fig. 46. Load vs. Midspan Deflection (Center Line) - Girder 3, Derailment Test.
...... (/)
0. ..ll:: -0 <t 0 .....J
w .....J X <t
o I I I ,..._ 5.0 x CLL I IOl
I 0 0 -
I
I
0 I{)
0
3.8 x CLL
,._/ 7 .-a
!:2L ✓ 2.8 x CLL
~ 2.0 x CLL
CLL
2 4
MIDSPAN DEFLECTION (in.)
YIELDING OF STIRRUPS
6
Fig. 47. Load vs. Midspan Deflection (Center-Line) - Girder 3, Derailment Test•
8
i SUPPORT
t L/6
t L/6
t L/6
- ---
"'--58.2k {~2 xCLL)
- - - ELASTIC THEORY (UN CRACKED)
TEST
i GIRDER l SUPPORT
t L/4
t L/4
t
C LL = CRUSH . LIVE LOAD
2.0
DEFLECTION { in. )
Fig. 48. Deflection Along Center-Line Axis - Girder 3, Derailment Test.
-fl) 0. ~ -0 <t 0 _J
w _J X <t
0 CX)
0 v
~H61
CLL
2.8 x CLL
2.0 x CLL
0 Fl RST CYCLE
~ SECOND ·CYCLE
CLL C CRUSH LIVE LOAD
~ I = 0.087'~ IMMEDIATELY AFTER UNLOADING
~2 = 0.057': AFTER I. 3 HOUR
1.0 2.0
MIDSPAN DEFLECTION (in) \
Fig. 49. ~oad vs. Midspan Deflection (Heavily-Loaded Stem) - Girder 3, .. Derailment Test.
3.0
0, I I I 5.0 X CLL _.e- I I() :,;;;;>"
0 0
-en a. ~ I ✓ 2.8 x CLL -0 <( 0 ...J
w I j 2 .0 x CLL ...J X <( 0
I()
CLL
0 2 4 6 8
MIDSPAN DEFLECTION (in.)
Fig. 509 Load vs. Midspan Deflection (Heavily-Loaded Stem) - Girder 3, Derailment Test.
t. SUPPORT L/ 6 L/6 L/6 t. GIRDER L/4 L/4 t. SUPPORT
t t t t t t
a1.ok (~ 2.8 x CLL)
CL L = CRUSH LIVE LOAD
2.0
DEFLECTION ( In.)
Fig. 51. Deflection Along Heavily-Loaded Stem - Girder 3, Derailment Test.
--Q)
~ I a. ~ -w ~ 0 a:: 0 t-
w _J X <(
or-----------r-------------..-----------..--------. ~
0 0 C\I
0 0
ELASTIC THEORY ( UNCRACKED ) ( 22
11 END DIAPH.)
2 .Q x CLL
CLL
0.2
MIDSPAN
2 .8 x CLL
C LL = CRUSH LIVE LOAD
0 .4 0.6
ROTATION (degree)
Fig. 52. Torque vs. Midspan Rotation - Girder 3, Derailment Test
0 - 0 - vi - 'YIELDING Q)
;/ -~ OF Q) - STIRRUPS I
0. .3' -IJJ ::> 0 a:: 0
I ""'/ _/' 2.8 x CLL t-
IJJ 0 .J X 0 ~ (\JI ff 2 x CLL
CLL CLL = CRUSH LIVE LOAD
0 2 3
MIDSPAN ROTATION (degree)
Fig. 53 Torque vs. Midspan Rotation - Girder 3, Derailment Test.
t. GIRDER 11 1-5 11 131-1 11 131-1 11 191-7 11 18
1-0
11
t t t t t t ..::: I =.;;M
........... --- - --- ----
172.9k(~ 2 x CLL) .
-----'---~
240.3k (~2.8x CLL)
CLL =
ROTATION (degree)
Fig. 54. Torsional Rotation Along Girder 3, Derailment Test.
ELASTIC THE ORY (UNCRACKED)
TEST
CRUSH LIVE LOAD
-(/)
0.. --~ -C)
<t 0 _J
w _J
X <t
0 LO
0 C\J -
0 r()
0
#84tJ #9
<!)
r z C) _J
w >-
20 40 60 20 40
STIRRUP STRESS ( k.s.i.)
Fig. 55. Stresses in the Outer Legs of Stirrup in the Heavily-Loaded
South Stem - Girder 3, Derailment Test.
<!) z C) _J w >-
60
-iJ,
Q.
.ll: -0 <( 0 _j
w _j X <(
~
0 <D
d I'()
0
#4
20 40
#9#6
(.!)
lJJ z 0 _j w >-
60 ·20 40
STIRRUP STRESS (k.s.i.) Fig. 56. Stresses in the Inner Legs of Stirrups in the Heavily-Loaded
South Stem - Girde r 3, Derailment Test.
(.!) z 0 _j w >-
60
-fl) C. ~ -0 < 0 _J
w _J X <
0 O')
0 <.D
0
""
0
*I
(!)
f z 0 _J w >-
20 40 60 20
STIRRUP STRESS ( k.s.i.)
Fig. 57 . Stresses in the Outer ~ egs of Stirrups in the
N-orth Stem - Girder 3, Derailment Test.
(!)
z 0 _J w >-
40 60
-,.,, Q.
.:it! -C)
<t 0 ...J
w ...J X <t
#3i4 42
Ji/
0 r<)
0 20
(!) z C) ...J w >-
40 60
STIRRUP STRESS (k.s.l.)
Fig. 58. Stirrup Stresses in the End Diaphragm - Girder 3,
Derailment .Test.
Fig. 59. Flexural Cracking Pattern - Girder 3 at
5.0 x crush Live Load (2.5 x Derailment Load)
Fig. 60. Shear Cracking Near Midspan - Girder 3, at
5.0 x Crush Live Load (2.5 x Derailment Load)
(a) At 2.0 x Crush Live Load
(b) At 5.0 x Crush Live Load
Fig. 61. Cracking Pattern at the Heavily-Loaded End - Girder 3
(a) (b)
Fig. 62 Crack Pattern at the Joint Between Stem and End Diaphragm - Girder 3,
at 5.0 x Crush Live Load (2.5 x Derailment Load)
Fig. 63. Crack Pattern of End Diaphragm - Girier 3 at
5.0 x Crush Live Load i2.5 x Derailment Load)
(a) East Face
(b) West Face
Fig. 64. Crack Pattern of Midspan Diaphragm - Girder 3 at
5.0 x Crush Live Load (2.5 x Derailment Load)
Fig. 65. Local Horizontal Cracks at Midspan - Girder 3 at
5.0 x Crush Live Load (2.5 x Derailment Load)
9'
G H
I 39
1- 3
11 39
1-3 11 9"
A. LONGITUDINAL . SECTION
t 1s'-011
------------- ---1------------
B. PLAN
BOLT
21~ 3
11-.8
11 STEEL BLOCK
5F08 3F25 4Fl6 3F25
C. CROSS SECTION D. CURB DETAILS
Fig. 66 Lo.:ation of Wheel Loads in Derailment Condition
..
Fig. 6 7. Instrumentation in Flange Slab Test
60.,_ ______________________________ __
50 3.26 x CLL (47.1 KIPS}
2.Sx CLL 40
-en Q.
~ -Q.V
30r 2.0 x CLL
.. 0 <t g ....J
20 w w ::x: 3:
~ 10 I / //
CLL = CRUSH LIVE LOAD
.05 .10 .15 .20
AV. DEFLECTION ( IN.)
Fig. 68 Load vs. Deflection Curve for Flange Slab Test
TOP SURFACE
EE
6
EDGE SURFACE
~ F - ----'LOCKOUT (J)
H3
~
7 6b-t 7
I~ EDGE -_Qf::._STEM"l_
3 _-J ~~ 7 6 6 ~ 16 7, 6 _ ~ ~ 14 i EE~ ~__14
V rzzm 14 rzm, Z"\1 &
'l_ LOAD POINT SECTION H
:-1\ ~~14 CURB EDGE OF FLANGE
WAD POINT SECTION G
TOP
BOTTOM SURFACE
16 \
6
EDGE OF STEM
Fig . 69 Crack Pattern After Failure at l.6 times Crush Live Load
Fig. 70. Punching Shear Failure at Load Point H
Fig. 71. Punching Shear Failure at Load Point G
-
