NZ Institute of Surveyors

Unbound Granular Pavement Unbound Granular Pavement DesignDesign

Tonkin & Taylor Ltd, Pavements GroupPavements Group

Coverage:g

Scope

ConceptsR h bili i Rehabilitation New pavementsAUSTROADSConventional/Mechanistic

Transit Research

Otago IssuesCase Histories

Notes and references available on www

Structural Design MethodsStructural Design Methods

1. Experience based1. Experience based 2. Empirical2. Empirical 3. Mechanistic3. Mechanistic--EmpiricalEmpiricalpp

H H 11

pp

HHSubgrade TypeSubgrade TypeTTH H 22H H 33VV

rraaff

StandardStandardThicknessesThicknessesfor variousfor variousff

ffii

for variousfor variousCombinationsCombinations

cc

DesignDesign

• Traffic

Experienced Based Mechanistic-Empirical•TrafficTraffic

• Materials• Long term in situ water content• Chart Design

•Traffic•Materials•Long term in situ water content•Design criteriaChart Design

- (Austroads Fig. 8.4)Design criteria

•Structural analyses•Determine stresses/strains•Compare design criteria

>>>New pavements

p g-(Layered Elastic Theory)

>>>Rehabilitation, widening, g

Post-construction verification: B/2 compliance, Deflection

Empirical Empirical –– Deflection based acceptanceDeflection based acceptance

Limitations Limitations -- Deflection Bowl ShapeDeflection Bowl Shape__

Highstrains Low

__strains

Distributed

Strains

Stress/strain –not deflectionnot deflectionDeflection cost effective on small projects, in “normal” terrain

Layered Elastic DesignLayered Elastic Design

Falling Weight Deflectometer

Layered Elastic DesignLayered Elastic Design

Falling Weight Deflectometer

•Sensors lowered onto existing pavement•Sensors lowered onto existing pavement•Loading plate with central sensor•Impact load 30 ms•Surface deflection bowl recordedSurface deflection bowl recorded•Buffer design developed that providesstresses and strains within the pavementthat correspond with those induced byp ya heavy vehicle (8 tonne axle)•Test is repeated to confirm accuracy

Layered Elastic DesignLayered Elastic DesignFWD RecordLoad and deflectionsVs TimeVs Time

Deflections from9 geophones from centref l d l t t 1 5of load plate to 1.5 m

Load pulse

< 30 milliseconds >

Structural InterpretationStructural Interpretation--back analysisback analysis

• Multi-layer elastic model, back-calculates moduli from deflection bowl:

EFROMD2EFROMD2ELMODEVERCALC (www)( )

Calculate stiffnesses of each layer from the impact load and shape of the deflection bowl using elastic theory

• Layer Stiffness = E x h3 .

Back analysis inputsBack analysis inputs

• FWD Impact Stress and Deflections• Pavement type (AC, chipseal, stabilised)• Layer thicknesses (up to 3, maybe 4 layers)

– as builts– maintenance records

t t it ( if iti ) – test pits (esp. if sensitive) • Environmental factors

temperature (AC only)– temperature (AC only)– seasonal effect.

BackBack--analysis outputsanalysis outputs

• Subgrade modulus (accurate)• Subgrade modulus non linearity (soil type/drainage)• Subgrade modulus non-linearity (soil type/drainage)• Layer moduli (E = 10 CBR approx.)

Typical values:

• M/4 basecourse 100 - 1000 MPa• AC 2000 - 6000 MPa• Cemented 2000 - 40,000 MPa

Structural Interpretation Structural Interpretation -- Forward AnalysisForward Analysisyy

CIRCLY, ODEMARK, ELMOD, EVERSTRESS etchttp://www.wsdot.wa.gov/TA/Software/p g

• Inputs– Pavement layer thicknesses y

Layer moduli (from back analysis)

• Deterioration models• Deterioration models– Established Strain Criteria .– AUSTROADS 92, AASHO Road Test– ARRB TR Model– ARRB TR Model– HDM

Mechanistic Analysis Strain CriteriaMechanistic Analysis Strain Criteria

10,000TRRLNottinghamA h lt I tit t

eg AUSTROADS:

ostr

ain

Asphalt InstituteNaasraAustroads 98TNZ PremiumBritish PortsShellAustroads Isotropic - Subgrade

Load Repetitions = [9300/microstrain]^7ESA - Equivalent Single Axles

1,000

ssib

le m

icro

p gAustroads Isotropic - Non subgradeLEDFAA 1.3

Per

mis

SAR: Standard Axle Repetitions calculated using damage exponent of

1001,000 10,000 100,000 1,000,000 10,000,000 100,000,000

Number of load repetitions

calculated using damage exponent of 4 (slope above) Chart Design

Subgrade Strain RatioSubgrade Strain Ratio

V i l i f b dVertical strain at top of subgrade________________________________

Allowable Strain (Austroads)

SSR>1 Premature distress likelySSR<1 Over designSSR<1 Over design SSR=1 Cost effective design

Structural Interpretation Structural Interpretation -- OutputsOutputs

• Basecourse strain• Subgrade strain • Subgrade strain ratio *• Critical layer (greatest strain) and distress mechanism• Depths of digouts or reconstruction• Depths of digouts or reconstruction.• Thickness of overlay required -5 options +

– AUSTROADS (GMP, Simplified 1992 & 2004)– TNZ Supp. (SHPDRM & Precedent Strain– AASHTO, TRRL

• Depth of cement stabilisation required (TNZ p q (Supplement)

• Residual Life

Subgrade Design CBRSubgrade Design CBR

• Precedent from as-builts of nearby roadsi it l t diti f b d id d- in situ long term condition of subgrade considered

• Laboratory CBR (soaked vs unsoaked)• ScalaScala• FWD – nearby surfaces, or can be directly on subgrade,

grass, loose gravel or cohesive soil with crust (elastic)

Subgrade Modulus, E (MPa) = 10 x CBR

• Austroads – 10%ile CBR • NB - Transit require 95%reliability

Otago FWD Survey - Cumulative Distributions

100

Camp Rd 0.932 - 1.4

Coast Rd 5.185 - 5.6

Exmouth St 0.000 -

Stirling St 0.000 - 0.

Allans Beach Rd 0.1

80

90

< X

Andersons Bay Rd L

Barr St 0.000 - 0.88

Birch St 0.110 - 0.28

Brockville Rd 0.412

Browns Rd 0.129 - 2

Allans Beach Rd 0.1

50

60

70

f sa

mpl

es Allans Beach Rd 0.1

Andersons Bay Rd L

Barr St 0.000 - 0.88

Birch St 0.110 - 0.28

Brockville Rd 0.412

Browns Rd 0.129 - 2

30

40

50

rcen

tage

of

Coast Rd 10.496 - 1

County Rd 0.175 - 0

Gladstone Road 3.0

Grove St 0.271 - 0.5

Harrington Point Rd

Highgate 0.358 - 0.9

10

20Per

g g

Highgate 2.401 - 3.0

Kaikorai Valley Rd 3

Larnach Rd 0.530 -

Macandrew Rd 1.28

Mitchell Ave 0.000 -

P k S 0 000 0 160

0 5 10 15 20

Austroads CBR

Park St 0.000 - 0.16

Shetland St 0.957 -

Signal Hill Rd 1.511

Station Rd 0.000 - 0

Station Rd Roundab

Wilmott St 0 000 0

80

90

100

X

60

70

80

mpl

es <

X

40

50

60

ge o

f sa

m

20

30

40

Per

cen

tag

Large range (almost an order)Long term in situ w/cDesign CBR usually 2 6

0

10

20P Design CBR usually 2-65 %ile CBR = 0.4 x Median10%ile CBR = 0.5 x Median

0

1 10 100

Austroads CBR

Traffic Traffic --ESAESA

Presumptive Traffic Load DistributionPresumptive Traffic Load DistributionTNZ Supplement to Austroads GuideTNZ Supplement to Austroads Guidewww.transitnz.govt.nz www.transitnz.govt.nz

S iS i i fi f-- State Highways State Highways -- Too conservative for local roadsToo conservative for local roadsAustroads 2004 Chapter 7Austroads 2004 Chapter 7

DesignDesign

• TrafficExperienced Based Mechanistic-Empirical

•Traffica c• Materials• Long term in situ water content• Chart Design

•Traffic•Materials•Long term in situ water content•Design criteriaChart Design

- (Austroads Fig. 8.4)Design criteria

•Structural analyses•Determine stresses/strains•Compare design criteria

>>>New pavements

p g-(Layered Elastic Theory)

>>>Rehabilitation, widening, g

Post-construction verification: B/2 compliance, Deflection

Transition to Mechanistic DesignTransition to Mechanistic Design

http://www.pavementanalysis.com/pages/Applications/PaveDesign/PaveDesign.php

Austroads Sub-layering

Design for CBR 2 (E=20 MPa)

SubSub--layer Modular Ratios layer Modular Ratios --UnboundUnbound

E base doubles for each 125 mmE i d d t f b d E base doubles for each 125 mmE base independent of subgrade

Austroads SubAustroads Sub--layer Modular Ratioslayer Modular RatiosAustroads (2004) Chapter 8•Divide the granular materials into 5 layers of equal thickness•Adopt the vertical modulus for the top sub-layer from:

E top of base = E subgrade x 2(total granular thickness/125)

(But not exceeding tabulated upper bounds for the materials.)(But not exceeding tabulated upper bounds for the materials.)

•Determine the modular ratio of successive sub-layers from:R = [E top of base / E subgrade ] 1/5

•Calculate the modulus of each layer beginning with that•Calculate the modulus of each layer beginning with that immediately overlying the subgrade of known modulus.In summary:E top of base = E subgrade x (function of total granular thickness)

With little dependence on the quality of the unbound granular materialif proper compactive effort is applied.

Modular RatiosModular Ratios

Transit – Research into Performance Based Specificationsusing FWD.

Normalised Modular Ratio: NMRNormalised Modular Ratio: NMR

NMR = As-built modular ratioStandard modular ratioStandard modular ratio

Measure of compaction compliance with Austroads expectations

Case Histories: PostCase Histories: Post--constructionconstruction

10000

100000

a)

Basecourse Modulus

Upper Subbase Modulus

Lower Subbase Modulus

Subgrade Modulus

1000

r Mod

ulus

(MP

a Normalised Modular Ratio

10

100Laye

10

0.5

1

1.5

2

2.5

NM

R

Moduli and normalised ratios for firm subgrade

0 0.5 1 1.5 2Station (km)

0

Case Histories: PostCase Histories: Post--constructionconstruction

10000

100000

a)

Basecourse Modulus

Upper Subbase Modulus

Lower Subbase Modulus

Subgrade Modulus

1000

er M

odul

us (M

Pa Normalised Modular Ratio

10

100Laye

10

0.5

1

1.5

2

2.5

NM

R

Moduli and normalised ratios for variable subgrade

0 1 2Station (km)

0

Case Histories: PostCase Histories: Post--constructionconstruction

100

Variable Subgrade CaseFirm Subgrade Case

75

ampl

es <

X

50

enta

ge o

f Sa

0

25

Per

c

Cumulative Normalised Modular Ratio Distributions for New Trafficked Pavements

0.5 1 1.5 2X axis - Normalised Modular Ratio

Transit LTPP Benchmark SitesTransit LTPP Benchmark Sites

Case Histories: Transit NetworkCase Histories: Transit Network100

X1999 All Benchmark Sites2000 All Benchmark Sites2001 All Benchmark Sites

50

75

Sam

ples

<

25

50

rcen

tage

of

0

25

Per

Cumulative distribution of normalised modular ratio for all Benchmark Sites

0 1 2 30.5 1.5 2.5X axis - Normalised Modular Ratio

Case History: ConstructionCase History: Construction

100

SH 1 Pre-Sealing after Additional CompactionSH 1 Pre-Sealing Initial Compaction

75

mpl

es <

X

50

enta

ge o

f Sam

0

25

Perc

e

Increase in Normalised Modular Ratio with Additional Compaction

0.5 1 1.5 2X axis - Normalised Modular Ratio

0

Construction QA Construction QA

Construction testing allows asConstruction testing allows as--constructed constructed modular ratios to be compared with expected modular ratios to be compared with expected results from known good practice, thereby giving results from known good practice, thereby giving

i di t f i di ti di t f i di tan immediate performance indicatoran immediate performance indicator

It provides a check for overall B/2 compliance as It provides a check for overall B/2 compliance as h f ll d h f i i l dh f ll d h f i i l dthe full depth of construction is evaluated, the full depth of construction is evaluated, not not

just the surface layerjust the surface layer

Allows decisions on severity of problem if B/2 Allows decisions on severity of problem if B/2 failsfails

Construction QAConstruction QA

The normalised modular ratio is a quantitative The normalised modular ratio is a quantitative measure of construction uniformity and stiffness ofmeasure of construction uniformity and stiffness ofmeasure of construction uniformity and stiffness of measure of construction uniformity and stiffness of the pavement layers, relative to the subgradethe pavement layers, relative to the subgrade

The performance indicator is independent ofThe performance indicator is independent ofThe performance indicator is independent of The performance indicator is independent of subgrade weaknesses or subgrade nonsubgrade weaknesses or subgrade non--uniformityuniformity

Readings should preferably be taken duringReadings should preferably be taken duringReadings should preferably be taken during Readings should preferably be taken during construction or prior to sealing or application of a construction or prior to sealing or application of a bound surface layerbound surface layeryy

Otago Issues: Very soft subgrades (CBR<3)Otago Issues: Very soft subgrades (CBR<3)• Austroads Light Traffic – 100-150 mm stabilised then CBR =3• Transit – 150+ mm stabilisation

d l- geogrid & geotextile *- 150 mm sacrificial granular layer and same CBR

• Geotextile – ideal if cost effectiveGeotextile ideal if cost effective• Geogrid – applicable in highly yielding subgrades (deflection

100 mm in trials). * Transit Supplement: Caution: “ It is up to the geosynthetic supplier to provide relevant and credible evidence that geosynthetic supplier to provide relevant and credible evidence that such savings are applicable for the particular product in question”

• Geogrid trials on SH1 Fairfield Bypass – evidence lackingG id d t t bl i hit & i l t t ti l• Geogrid undetectable in hit & miss layout –construction only

• Drainage plus validation of design CBR for any SIL on CBR<3

Otago Issues: SchistOtago Issues: Schist--derived siltderived silt

• Transit Supplement: Care should be taken assessing silty and sensitive subgrades. They can be significantly weakened by the i i t f t ti i t d thi inappropriate use of construction equipment and this should be noted in the contract documents.

• Otago schist deri ed micaceous silts• Otago schist-derived micaceous silts:+1 m of granular subbase needed on SH 1

pavements with silt fills compacted wet of pavements with silt fills compacted wet of optimum.

Otago Issues: Solid DensityOtago Issues: Solid Density• B/2 Compliance• AP65- Correction for oversize in %MDD • Saturation and %MDD are very dependent on

solid density• Solid density often “assumed”• Percentage Solid Density Concept – an

independent check on B/2 resultsindependent check on B/2 resultsGrading exponent “n”

Fine M/4 n=0 41 coarse M/4 n= 63Fine M/4, n=0.41, coarse M/4 n=.63Dmax % = 86 + 20(0.5-n)

Otago Issues: FrostOtago Issues: Frost

• Transit Supplement: Indefinite guidance.• Frost Resistant Design and Construction of Pavements • Frost Resistant Design and Construction of Pavements

in Central Otago www.pavementanalysis.com

S d E i l t > 50 • Sand Equivalent > 50 or;• Percent passing 2.36 and 0.15 mm sieves give

P 2.36 / P 0.15 > 3 or;.36 0. 5• Freeze-thaw Heave < 14 mm

Laboratory after compaction or better, field afterLaboratory after compaction or better, field aftercompaction

Case Histories

Demonstrating achievement of Demonstrating achievement of intended design life at end of constructionintended design life at end of construction

Rehabilitation projectsRehabilitation projects

Long term study: Waihola passing laneLong term study: Waihola passing lane

New Construction

New ConstructionPrePre--seal Postseal Post--sealseal

Design AssuranceDesign Assurance

Deflection Surveys Before and After Construction

02 0 04 0 0 M /4 O ve rlay (m m )

0

1 0 0 0 0 1 00 0

02 0 04 0 0

S ta b ilisa tion D e p th (m m )

C B R

B a s e co u rse M o d u lu s (M P a )

1 0 0

1 0 0 0

5 00

1 0 0

5 0

S u b g ra d e M o d u lu s (M P a )1 0

1 0 0 1 0

5

0 0 0 2 0 4 0 6 0 8 1 0 1 2 1 4 1 6 1 8 2 0 2 2012

S u bg ra de S tra in R a tio

0 .0 0 .2 0 .4 0 .6 0 .8 1 .0 1 .2 1 .4 1 .6 1 .8 2 .0 2 .2 C h a in a g e (k m )

Before After M/4 overlay

200

400 M/4 Overlay (mm)200400 M/4 Overlay (mm)

0

10000 1000

0200400

Stabilisation Depth (mm)

CBR

0

10000 1000

0200400

Stabilisation Depth (mm)

CBR

Basecourse Modulus (MPa)

100

1000

10000500

10050

10

Basecourse Modulus (MPa)

100

1000

10000500

100

50

10

Subgrade Modulus (MPa)10

5

2Subgrade Strain Ratio

Subgrade Modulus (MPa)10

5

2Subgrade Strain Ratio

8.2 8.4 8.6 8.8 Chainage (km)

012

8.2 8.4 8.6 8.8 Chainage (km)

012

Construction verification on yielding subgradeBefore After part M/4 overlay

Determine achievement of objective (full design life)

Di ti i h f d fi iDistinguish cause of any deficiency:- Design issue (over design / under design)

Construction QA- Construction QA- Subgrade non-uniformitySubgrade strain ratioSubgrade strain ratio

(Actual / Austroads allowable strain)

References & FiguresReferences & Figures

• References, 2004 anb LVR, both TNZ supplements to Austroads 2005 and 2002? **• Light Pavements – corresponding figure with same reliability No• Transit requirements 95% 25 year design (but note 10 percentile CBR normally q y g ( p y• dcc criterion is for acceptance of new pavements• Austroads 2004 – Fig 8.4 with hatched zone from S4 – reliability=?• Traffic p 32 of TNZ supp to 04, ESA=1.4*HV, 3% growth, 25 years **• Selection of design CBR, (copy ex Aust2004), Scala (silts not sands), deflection testing, design moisture content. Back

l l t d CBR (P d t th d)calculated CBR (Precedent method).• Mechanistic design concepts• Subgrade strain criterion – all graphs from Odemark 2006*• Determination of pavement parameters – using FWD – BMS subsection Es cumulative• Manapouri es plotsp p• Mechanistic design spreadsheet – not if Fig 8.4• Field verification of subgrade – during construction – Scala, FWD • Weaving excess moisture.• Compliance monitoring- spreadsheet ‘refer to nmr if fails

f t it i i f **• frost criterion paper in references **• Verification of design life – post construction testing• Rehabilitation Design – Transit supplement

Subgrade improvementSubgrade improvement

• SIL propertiesp p• Schist derived subgrade issues• Geogrid/geofabricGeog d/geo ab c

Acceptance testing of pavement layersAcceptance testing of pavement layers

• Austroads moduli for unbound layersy• Checks on compaction• Case Histories from Otagog

Demonstrating achievement of intended Demonstrating achievement of intended d i lif d f id i lif d f idesign life at end of constructiondesign life at end of construction

• Transit requirements• Transit requirements• Local Authority requirements• Case History from Otago• Case History from Otago