4/28/2016 1 Mark H Wayne, Ph.D., P.E. Director of Application Technology Kent Seminar Series University of Illinois, Urbana-Champaign April 21, 2016 Mechanical Stabilization of Unbound Layers and Incorporation of Benefits in AASHTO ‘93 and M-E Analysis of Flexible Pavements Lecture Outline Tensar International Stabilization Function & Confirmation Through Research AASHTO Empirical Approach Mechanistic-Empirical Approach New Pavement Performance Evaluation Technologies
Transcript
4/28/2016
1
Mark H Wayne, Ph.D., P.E.Director of Application Technology
Kent Seminar SeriesUniversity of Illinois, Urbana-ChampaignApril 21, 2016
Mechanical Stabilization of Unbound Layers andIncorporation of Benefits in AASHTO ‘93 and M-E
Analysis of Flexible Pavements
Lecture Outline
Tensar International
Stabilization Function & Confirmation Through Research
AASHTO Empirical Approach
Mechanistic-Empirical Approach
New Pavement Performance Evaluation Technologies
4/28/2016
2
Tensar Corporation is the parent company of severalwholly-owned, market-leading subsidiaries including:
• Tensar International Corporation
• Geopier Foundation Company
• North American Green
Tensar Group Overview
Tensar International
4/28/2016
3
“Everything From the Ground Down”
ReinforcedSlope
RetainingWall
EmbankmentStabilization
RoadSubgrade
StabilizationPavement
Optimization
Manufacturing
4/28/2016
4
4/28/2016
5
Lecture Outline
Tensar International
Stabilization Function & Confirmation Through Research
Proposed Definition by ISO TC221 - WG2
Stabilization: Improvement of the mechanicalproperties of an unbound granular material byincluding one or more geosynthetic layers suchthat the deformation under applied loads isreduced by minimizing soil particle movement.
Mechanical Stabilization is a more appropriate description –distinguishes from Chemical Stabilization, Lime Stabilization andothers
What is Stabilization?
4/28/2016
6
Importance of Stabilization
A half section of a typical railroad track structure was constructed. TriAx TX190L geogrid was installed 10” below the top of the
ballast. SmartRock is installed above geogrid and record real-time particle
movement including translation and rotation.
Particle Movement inside Railroad Ballast
Presented at the 2016 TRB conference,“Effect of Geogrid on Railroad Ballast Studied by SMART ROCK”Liu, S., Huang, Hai, Qiu, T. and Kwon, J.
4/28/2016
7
Research: Real Time Rotation
Rotation + Translation
4/28/2016
8
Laboratory setup
PARTICLE TRANSLATIONAL MOVEMENT was significantlyreduced with the inclusion of TX190L geogrid.
Particle Movement inside Railroad Ballast
4/28/2016
9
PARTICLE ROTATION was significantly reduced with the inclusionof TX190L geogrid.
Particle Movement inside Railroad Ballast
WITHOUT Geogrid WITH Geogrid
Visualized motion of SmartRock in ballast
Presented at TRB2016 conference,“Effect of Geogrid on Railroad Ballast Studied by SMART ROCK”
4/28/2016
10
Multi-Level Shear Box Testing – with Geogrid Shear plane 1 – top of the sand layer Shear plane 2 – 100mm above top of the sand Shear plane 3 – 200mm above top of the sand Shear plane 4 – 300mm above top of the sand
Multi-level Shear Box
The geogrid in the ballast layer increased the peak shear force atall of the four levels. The shear force increase is a true indication of the effect of
aggregate confinement.
Shear Force at Various Distances
4/28/2016
11
Large Scale TriAxial Testing
The University of Illinois Triaxial Ballast Tester or TX-24 Specimen Size: 12” x 24”
10 wheel crossings (500 N, 0.5 m/s) 5 kPa normal stress is applied on load walls during the test
ITASCA DEM - Moving wheel load simulation
wheelcycles backand forth
non-stabilised- 9th run
mechanically stabilised- 9th run
4/28/2016
13
• DEM
Forces in the Geogrid Under a Wheel Loading
yz
xSS20 9th run
Fmax = 33.6 lb/ft
TX160 9th run
Fmax = 18.5 lb/ft
4/28/2016
14
ITASCA DEM - Lateral and Vertical Confinement
Biaxial Geogrid = reducedvertical and horizontaldisplacement versus control
TriAx = significantly less verticaland horizontal displacementversus control and biaxialgeogrid. Maintain particleshape and position = longterm stiffness retention!
(Particle Movement OverTime! = reduction in layerstiffness over time!)
Stabilization/Reinforcement Functions
Geogrid orGeosyntheticwhere particleconfinement isnot developed
Reinforcement
Geogrid whereinterlockresults inefficientparticle
confinement
Stabilization
4/28/2016
15
TRL Trafficking - Jenner, Watts & Blackman (2002)
Investigating different forms of geosynthetic Soft subgrade approx. 2% CBR 9,000 lb wheel (equal to 1 ESAL) Surface rut depth and deformation measured Subgrade profile measured after exhumation
Trafficking – 10,000 passes
4/28/2016
16
-140
-120
-100
-80
-60
-40
-20
00 2000 4000 6000 8000 10000
Mea
n r
ut
dep
th (
mm
)Passes
Membrane
Confinement
Control
Membrane ConfinementControl
-0.4
-0.3
-0.2
-0.1
0
0.1
0 0.4 0.8 1.2 1.6 2 2.4
Dep
th b
elo
w d
atu
m (
m)
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)
N = 0
-140
-120
-100
-80
-60
-40
-20
00 2000 4000 6000 8000 10000
Mea
n r
ut
dep
th (
mm
)
Passes
Membrane
Confinement
Control
Membrane ConfinementControl
-0.4
-0.3
-0.2
-0.1
0
0.1
0 0.4 0.8 1.2 1.6 2 2.4
Dep
th b
elo
w d
atu
m (
m)
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)
N = 100
4/28/2016
17
-140
-120
-100
-80
-60
-40
-20
00 2000 4000 6000 8000 10000
Mea
n r
ut
dep
th (
mm
)Passes
Membrane
Confinement
Control
Membrane ConfinementControl
-0.4
-0.3
-0.2
-0.1
0
0.1
0 0.4 0.8 1.2 1.6 2 2.4
Dep
th b
elo
w d
atu
m (
m)
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)
N = 200
-140
-120
-100
-80
-60
-40
-20
00 2000 4000 6000 8000 10000
Mea
n r
ut
dep
th (
mm
)
Passes
Membrane
Confinement
Control
Membrane ConfinementControl
-0.4
-0.3
-0.2
-0.1
0
0.1
0 0.4 0.8 1.2 1.6 2 2.4
Dep
th b
elo
w d
atu
m (
m)
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)
N = 500
4/28/2016
18
-140
-120
-100
-80
-60
-40
-20
00 2000 4000 6000 8000 10000
Mea
n r
ut
dep
th (
mm
)Passes
Membrane
Confinement
Control
Membrane ConfinementControl
-0.4
-0.3
-0.2
-0.1
0
0.1
0 0.4 0.8 1.2 1.6 2 2.4
Dep
th b
elo
w d
atu
m (
m)
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)
N = 1,000
-140
-120
-100
-80
-60
-40
-20
00 2000 4000 6000 8000 10000
Mea
n r
ut
dep
th (
mm
)
Passes
Membrane
Confinement
Control
Membrane ConfinementControl
-0.4
-0.3
-0.2
-0.1
0
0.1
0 0.4 0.8 1.2 1.6 2 2.4
Dep
th b
elo
w d
atu
m (
m)
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)
N = 2,000
2000 passes
4/28/2016
19
-140
-120
-100
-80
-60
-40
-20
00 2000 4000 6000 8000 10000
Mea
n r
ut
dep
th (
mm
)Passes
Membrane
Confinement
Control
Membrane ConfinementControl
-0.4
-0.3
-0.2
-0.1
0
0.1
0 0.4 0.8 1.2 1.6 2 2.4
Dep
th b
elo
w d
atu
m (
m)
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)
N = 5,000
2000 passes
-140
-120
-100
-80
-60
-40
-20
00 2000 4000 6000 8000 10000
Mea
n r
ut
dep
th (
mm
)
Passes
Membrane
Confinement
Control
Membrane ConfinementControl
-0.4
-0.3
-0.2
-0.1
0
0.1
0 0.4 0.8 1.2 1.6 2 2.4
Dep
th b
elo
w d
atu
m (
m)
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)
N = 9,000
Subgradeprofile
2000 passes 9000 passes
4/28/2016
20
Membrane ConfinementControl
-0.4
-0.3
-0.2
-0.1
0
0.1
0 0.4 0.8 1.2 1.6 2 2.4
Dep
th b
elo
w d
atu
m (
m)
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)0 0.4 0.8 1.2 1.6 2 2.4
Distance across section (m)
N = 10,000
Subgradeprofile
2000 passes 9000 passes 10,000 passes
-140
-120
-100
-80
-60
-40
-20
00 2000 4000 6000 8000 10000
Mea
n r
ut
dep
th (
mm
)Passes
Membrane
Confinement
Control
Confinement
Tensioned membrane
Note:
The membranegeosynthetic hastwice the strengthof theconfinementgeosynthetic
Geosynthetic Functions - Permanent Roadways
Filtration Separation Reinforcement Stabilization
4/28/2016
21
Stabilization
Geogrid aperture size relative to aggregate size and grading FHWA Guideline: D50<Aperture Size<2D85
where “D” values are for aggregate placed on the geogrid.
Separation Check Piping Ratio = D15fill/D85subgrade <5 Average Size Ratio = D50fill/D50subgrade < 25
Lecture Outline
Tensar International
Stabilization Function & Confirmation Through Research
AASHTO Empirical Approach
4/28/2016
22
AASHO Road Test (late 1950’s)
One Subgrade Type…
A-6 / A-7-6 (Clay)Poor Drainage
4/28/2016
23
Controlled Construction Methods...
1950s’ Vehicle Loads...
4/28/2016
24
AASHTO Pavement Design Guide
Empirical methodology
Based on AASHO Road Test
Several versions: 1961 (Interim Guide), 1972,
1986, 1993
1986 Guide highlights need formechanistic design
Benefit of includinggeosynthetics in pavement isrecognised to: Improved life Reduced thickness
Benefits cannot be derivedtheoretically
Designs not easily translatedto other geosynthetics
Test sections are necessary toobtain benefit quantification
Users are encouraged toaffirm their designs with fieldverification
AASHTO: R50-09
4/28/2016
25
Full Scale Evaluation
USCOE Full Scale APT Studies
Accelerated Pavement Testing:
Provide full-scale pavement performancedata for TriAx for base enhancement designfollowing AASHTO '93 and/or M-Eapproaches.
Pavement layersrepresented by theirstructural number SN
Pavement conditiongiven by its presentserviceability indexPSI (p)
Traffic given by number of18 kip (80kN) ESA W18
Design with a Mechanically Stabilised Layer
SNmsl
SNmsl
Pavement Optimization Summary
Original DesignLife
6 X OriginalDesign Life
Original DesignLife, Lowest
First Cost
3 X OriginalDesign Life,Same Cost
4”
10”
3”
8”
3.5”
9”
4”
10”
4/28/2016
37
Pavement Optimisation – an existing proposal prior to optimisation
TraditionalPavementPa
vem
entq
ualit
y,Sa
lvag
e Va
lue
Serv
icea
bilit
y
Time
New or Reconstructed Pavement
Full
Rec
onst
ruct
ion
Req
uire
d
The Development of a Value Proposition
p0 = 5 for perfect pavement(this can never be attained)
p0 = 5
pt = 2
Original DesignLife
4”
10”
Pavement Optimisation – a short term value proposition approach• focus on the construction phase
Reduce the pavement to its optimum (thinnest) thickness, whilst retaining existing capacity
TraditionalPavement Time
New or Reconstructed Pavement
Full
Rec
onst
ruct
ion
Req
uire
d?
The Development of a Value Proposition
Original DesignLife, Lowest First
Cost
TensarPavement
Pave
men
tqua
lity,
Salv
age
Valu
e
Serv
icea
bilit
y
3”
8”
4/28/2016
38
Pavement Optimisation – a medium term value proposition approach• focus on the construction phase along with enhanced risk management benefits
Reduce the pavement thickness, whilst increasing the performance
TraditionalPavement Time
New or Reconstructed Pavement
Onl
y P
art
Rec
onst
ruct
ion
Req
uire
d
The Development of a Value Proposition
TensarPavement
3 X OriginalDesign Life,Same Cost Extended life = Reduced Costs
Pave
men
tqua
lity,
Salv
age
Valu
e
Serv
icea
bilit
y3.5”
9”
Pavement Optimisation – a long term value proposition approach• focus on the whole life cycle for the whole pavement structure
Maintain the pavement thickness, whilst increasing the whole life design capacity
TraditionalPavement Time
New or Reconstructed Pavement
Onl
y S
urfa
ceC
ours
e Tr
eatm
ent
Req
uire
dThe Development of a Value Proposition
TensarPavement
Extended life = Reduced Costs6 X OriginalDesign Life
Pave
men
tqua
lity,
Salv
age
Valu
e
Serv
icea
bilit
y
4”
10”
4/28/2016
39
Lecture Outline
Tensar International
Stabilization Function & Confirmation Through Research
AASHTO Empirical Approach
Mechanistic-Empirical Approach
Incorporating the geogrid effect into M-E Analysis
User Input
MechanisticAnalysis
TransferFunction
LifeEstimation
MaterialsClimateTraffic
Geogrideffect ondeterioration
Geogrideffect onmodulus
Life shiftfactors
Layeredelasticanalysis
4/28/2016
40
Mechanistic Empirical
1 – S Input
4 – A Input
3 – UG InputTriAx
2 – UG InputTriAx
EnhancedModulus
LEATransferFunction
ShiftFactor
TransferFunction
TransferFunction
TransferFunction
EnhancedModulus
LayerProperties
LayerProperties
Layer 3Life
Layer 4Life
Layer 2Life
Layer 1Life
Target ESALs
Me
et
Me
et
Me
et
Me
et
Incorporating the geogrid effect into M-E Analysis
Experts in the industry ofpavement design. Developed AASHTOWare
Pavement ME design softwareused throughout NorthAmerica today Currently Perform M-E
Validation and Calibration fornumerous State Department ofTransportation
Mechanistic-Empirical Analysis
4/28/2016
41
Lecture Outline
Tensar International
Stabilization Function & Confirmation Through Research
AASHTO Empirical Approach
Mechanistic-Empirical Approach
New Pavement Performance Evaluation Technologies
Elastic versus resilient modulus
Mr = (1-v2) f σo (a / dr)dr = recoverable deformation
E = (1-v2) f σo (a / d0)do = Elastic deformation
4/28/2016
42
Center for EarthworksEngineering Research
Assessment of Pavement Foundation Stiffness usingCyclic Plate Load Test, Presented by Mark H. Wayne
• Influence of load cycles
In-situ Resilient Modulus
Number fo Load Cycles
0 50 100 150 200 250
In-s
itu R
esili
ent M
odul
us (M
Pa)
60
80
100
120
140
160
180
Number of Load Cycles
0 50 100 150 200 250
Per
man
ent D
efor
mat
ion
(mm
)
2
4
6
8
10
12
14
16
18
20
TX160BX1200Control
35 to 345 kPa cyclic stress
GG2 TXGG1 BXControl
4/28/2016
43
dp = CNd
A power model describes the permanentdeformation versus load cycles responseto provide deformation forecastingcomparisons.
Monismith et al. (1975) described thepower model relationship for relatingpermanent strain to cycle loadings.
Post-compaction permanent strain isa function of the shear stressmagnitude and can reach anequilibrium state following the“shakedown” concept (see Dawsonand Feller 1999).
Number of load cycles, N
Perm
anen
t def
orm
atio
n,δ p
Weak Layer
Stabilized Layer(lower qualityaggregate)
Stabilized Layer(higher qualityaggregate)
f (material type, physicalstate, and stress conditions,Li and Selig 1994)
f (shear stress magnitude,aggregate abrasion resistance,resiliency of stabilizer)
Ingios 2-layer testing to determine baseand subgrade layer moduli values
4/28/2016
44
Two-Layered Analysis using Odemark’smethod of equivalent thickness concept
σo
Mr1, v1
Mr2, v2
dr,0
hdr,h
σo
Mr2, v2
Mr2, v2
he
e
dr,0
dr,h
Illustration of Odemark’s Method of Equivalent Thickness (MET) concept.
3222
211
)1(
)1(
vM
vMhh
r
re
'r,r)sg(r 'r
P)(M
21
)base(r
)sg(r
)base(r)sg(r
c M
r
h
)v(M
)v(M
r
hM
rf)(
2
2
322
21
02
1
11
1
11
11
Calculating Base and Subgrade LayerModulus – (AASHTO 1993)
“For the 10,000 cycle test,the in-situ resilient modulusrapidly increased in theaggregate base layer forthe first ~3000 cycles andthen continued to increaseat a slower rate. Based on apermanent deformationrate of 0.0001in./cycle thetransition from plasticdeformation accumulationto near-linear elastic occursat N* = 8,696 cycles. AtN*, the in-situ Mr wasabout 321,881 psi (2xhigher than the averagevalue from the 1000 cycletests).”
Testing ConductedMr of the mechanically stabilized base courseMr of the subgradeMr composite modulusModulus of subgrade reaction (k)Ev1 and Ev2 strain modulus testingResilient deflections (scaling exponent)
