VT/FHWA Center of Excellence for Polymer Composites

1
Development of Product Design Guides
John J. Lesko Thomas E. Cousins, Department of
Engineering Science Mechanics Department of
Civil and Environmental Engineering, Virginia
Tech Blacksburg, VA 24061 Dan E. Witcher
Glenn P. Barefoot Strongwell, Corp Bristol, VA,
24203
2003 Technical Conference on Construction,
Corrosion and Infrastructure Las Vegas, NV,
22-25 April 2003
2
Where is Virginia Tech?
Hawthorne St. Bridge
That Other Univ.
Troutville Weigh Station
Toms Creek Bridge
Tangier Island
Dickey Creek Bridge
Rt. 601 Bridge
3
Barriers to Routine use of FRP

• Administrative
• Fragmentation of the industry
• Lack of interdisciplinary training for design
  engineers
• Cost (first vs. life cycle)
• Limited commercial capital for development of new
  systems
• Incremental - piecemeal approach to FRP
  implementation in designs

• Technical
• Lack of design specifications
• Performance vs. Material specification
• Lack of sufficient long-term data and experience
• Plethora of new materials, additives
  combinations
• One-for-One material substitution
• Lack of confidence in adhesive bonding

4
Barriers to Routine use of FRP

• Administrative
• Fragmentation of the industry
• Lack of interdisciplinary training for design
  engineers
• Cost (first vs. life cycle)
• Limited commercial capital for development of new
  systems
• Incremental - piecemeal approach to FRP
  implementation in designs

• Technical
• Lack of design specifications
• Performance vs. Material specification
• Lack of sufficient long-term data and experience
• Plethora of new materials, additives
  combinations
• One-for-One material substitution
• Lack of confidence in adhesive bonding

5
Barriers to Routine use of FRP

• Administrative
• Fragmentation of the industry
• Lack of interdisciplinary training for design
  engineers
• Cost (first vs. life cycle)
• Limited commercial capital for development of new
  systems
• Incremental - piecemeal approach to FRP
  implementation in designs

• Technical
• Lack of design specifications
• Performance vs. Material specification
• Lack of sufficient long-term data and experience
• Plethora of new materials, additives
  combinations
• One-for-One material substitution
• Lack of confidence in adhesive bonding

6
Barriers to Routine use of FRP

• Administrative
• Fragmentation of the industry
• Lack of interdisciplinary training for design
  engineers
• Cost (first vs. life cycle)
• Limited commercial capital for development of new
  systems
• Incremental - piecemeal approach to FRP
  implementation in designs

• Technical
• Lack of design specifications
• Performance vs. Material specification
• Lack of sufficient long-term data and experience
• Plethora of new materials, additives
  combinations
• One-for-One material substitution
• Lack of confidence in adhesive bonding

7
Mil Handbook 17
Mission Statement Develop world-class engineering
handbooks for structural applications of
composite materials. These handbooks will include
standards for test/characterization methods,
statistics and databases, as well as guidelines
for processing, design and analysis.
http//www.mil17.org/
8
FRP Double Web Beam

• Pultruded by Strongwell, Corp.
• Both Hybrid and All Glass
• Hybrid, Non-Symmetric Layup
• Vinyl Ester Resin
• Glass 0, 45 90 plies
• Hybrid Carbon in top and bottom flanges
• Designed for the bridge, offshore and other

9
Modified LRFD Approach

• 8 36 DWB Design Guide
• Deflection (AB Allowables)
• Strength (AB Allowables)
• Stability
• Bearing
• Connections
• Fatigue Long Term

Reliability based approach to assessing A B
basis Allowables, as described through Weibull
Statistics
10
Extren 8 Double Web Beam (DWB) Characteristics
Pultruded Hybrid Glass Carbon/Vinyl Ester
Shape Ixx 129 in4 Iyy 31.8 in4 Sx 32.2 in3
Sy 10.6 in3 rx 3.07 in ry 1.52 in A
13.7 in2 A2 webs 5.36 in2 A2 flanges 7.44
in2 Weight/foot 11lbs/ft
11
Extren 36 Double Web Beam (DWB) Characteristics
Pultruded Hybrid Glass Carbon/Vinyl Ester
Shape Ixx 15291 in4 Sx 849 in3 Iyy 2626
in4 Sy 292 in3 rx 12.9 in ry 5.37 in A2w
50.1 in2 Asf 34.0 in2 Weight/foot 75
lbs./foot
Dimensions in inches
12
Load Resistance Factor Design (LRFD)
Resistance, R
Load, Q
AASHTO (1998) LRFD based design
13
DWB Design Guide Approach
User supplies loads and level of acceptable risk
based on change in Resistance
Cumulative Probability
14
Design Guide

• Material Specification
• Weibull Statistics Reliability
• 8 36 Deep DWB Design Guide
• Deflection (AB Allowables)
• Strength (AB Allowables)
• Stability
• Bearing
• Connections
• Fatigue Long-term

15
Stiffness Capacity
Unsupported Spans 8, 14, and 20
feet Instrumentation S Shear, B Bending, D
- Deflection
16
8 DWB
36 DWB
Top flange failure due to delamination
Support failures at spans shorter than 30
17
Moment Capacity Material
Moment capacity controlled by carbon/glass
interlaminar interface
18
DWB Strength vs. Span
19
Shear Deformation 8 DWB
A-basis Allowables E 5.66 Msi kGA 1.8 Msi-in2
20
Shear Deformation 36 DWB
B-basis Allowable E 6.1 Msi
21
Assessing Stiffness Allowables
Bending modulus is determined in the constant
moment section from axial strains
kGzyAv is not independent of Ezz And is therefore
determined from deflection under the 4 point
loading condition
22
Development of Allowables
Using Median Rank and Weibull statistics we
develop a and b
Establish the blower (5 confidence interval) and
the A and B basis allowables based on the desired
reliability
A Allowable
B Allowable
23
FRP Design Allowables
Weibull Cumulative Probability
Design Stress
24
Weibull Statistics on Modulus, Ezz
25
Design Tables
Backing out the allowable elastic shear
properties (kGzyA)A from a distributed loading
case
Developed effective allowable distributed loads
form Ezz and kGzyAv for a given dL/k
Allowable distributed loads, wA were then used to
compute (kGzyA)A
26
8 DWB Hybrid Beam A-basis Allowables
Ezz 5.66 x 106 psi kGzyAV 1.8 x 106 psi-in2
Mmax 36.1 kip-ft.
 
Based on a simply supported beam under
distributed load
Shear Deflection 30 _at_ 8 7 _at_ 20
27
36 DWB Hybrid Beam B-basis Allowables
Ezz 6.10 x 106 psi kGzyAV 46.2 x 106
psi-in2 Mmax 1139 kip-ft _at_30 Span
916.7 kip-ft 40-60 Span
 
Based on a simply supported beam under
distributed load
Shear Deflection 15 _at_ 30 5 _at_ 60
28
Lateral Torsional Buckling 8 DWB
Rotation lateral displacement allowed at
mid-span
Unsupported Spans Tested 40, 36, 32, 28, 24, 20
feet To L/90 No LTB observed! Impose Limit L/180
29
Lateral Torsional Buckling 36 DWB
Rotation lateral displacement allowed at
mid-span
Unsupported Spans Tested 60 feet To L/180 No
LTB observed! Impose Limit L/360
30
Bearing Capacity Web Buckling
Bearing controlled by web buckling
Factor of Safety
Total allowed bearing load
31
Bolt Bearing in Connections
Bolt bearing controlled by ultimate bearing
strength of the web material
FpCr 30 ksi
Allowable pin bearing load
Fastener edge distances 2 diameters or 1 (25mm)
minimum, which ever is greater. Fastener
pitch4 diameters or 3 (76mm) minimum, which
ever is greater.
32
So what about durability?
33
Toms Creek Bridge
Toms Creek Bridge
34
Toms Creek Bridge, June 1997
Deflection L/490 Wheel Load Distribution Factor
0.101 Dynamic Load Allowance 0.9
35
Beam Removal Replacement
Two beams having seen 15 months (Sept 1998) of
service were removed to assess remaining strength
and stiffness

• After 15 months of service...
• No residual creep deflection
• No reduction in residual strength and stiffness

36
Dickey Creek Bridge
Dickey Creek Bridge
37
Dickey Creek Bridge
Deflection L/1100 Wheel Load Distribution
Factor 0.2 Dynamic Load Allowance 0.36
38
How Does the Resistance Change?
FRP Life prediction is required as a function of
load and environmental history to assess the
changes in Resistance
Emphasis on Combined Environments CERF/MDA
Durability Gap Analysis
39
Developing Guidance on f
(FORM) Hasofer Lind (1974)
Resistance, R
Load, Q
CDF
PDF
Initial Resistance
Residual Resistance X-years of service
40
Estimating Remaining Strength Stiffness
FRP composites durability is best described by
nonlinear cumulative damage approaches where
residual strength and stiffness are tracked
during life
Degradation Processes

• Cycle dependent damage
• Kinetic
• Chemical
• Thermodynamic

Geometry Constitutive
Remaining Strength Of Critical Element
Initial Strength
Stress or Strength
Life
N
Stress on Critical Element
Reifsnider et al. (1975- present)
41
Simulation Approach
Loads

• Develop estimate on resistance based on stress
  analysis/material
• Develop load/environment history based on
  statistical description (Monte Carlo Simulation)

Yes
No
?
Compute f Pf
Life Prediction

• Input material characteristics (S-N curve,
  stiffness and strength reduction as a function of
  environment -including statistical description)

42
Engineering Practice Example
Element FRP Pultruded hybrid vinyl ester
structural girder Region Northeast US
(Environmental factors - thermal, moisture
UV) Rn X moment capacity (Resistance and
inherent resistance variation) gQn Operating
Moment based on stress analysis from AASHTO HS20,
with ADT 10,000, 30 fully loaded (Load and load
variation)
Example NOT meant for design
43
Conclusions

• FRP design guide
• Materials specification
• Laboratory testing
• Reliability based
• Long-term validity of as received design
  allowables
• Durability Modified LRFD for FRP a possibly
  means to gain acceptance among practicing
  engineers
• Suggest inspection cycle
• Phenomenological vs. First Principles
• Material Specification?
• Is this realistic or just academic???

44
Acknowledgements

• FHWA, Innovative Bridge Research Construction
  Program
• Virginia Dept of Transportation
• National Science Foundation
• Virginia Center For Innovative Technology
• Strongwell, Corp.

45
QUESTIONS?