Street Damage Restoration Fee StudyTesting Portion

^ty■LA* iu- '*4*rt__

l£

City of Los AngelesDepartment of General Services

Bureau of Street Services

Shahin and AssociatesMarch 2017

Form GS:STD 911CITY OF LOS ANGELES

DEPARTMENT OF GENERAL SERVICES STANDARDS DIVISION

2319 DORRIS PLACE LOS ANGELES, CA 9003!

(213) 485-2242 FAX (213) 485-5075

Lab No.: 17-702-1703 to 1780 17-741-031 to 264

Rec'd: 10/03/16 Rep’d: 03/24/17

Mr. Nazario Sauceda, Director Public Works / Bureau of Street Services

Keith Mozee, Assistant Director Steve Chan, BSS Pavement Management Dr. Mo Shahin, BSS Consultant Dr. James Crovetti

Attn:Cc:

Test Report ofLos Angeles Street Damage Restoration Fee (SDRF) Update Study

At the request of the Bureau of Street Services. Standards Division of the Department of General Services conducted pavement testing, analysis and overlay designs for the Los Angeles Street Damage Restoration Fee Update Study. The project was developed and conducted under the guidance of Dr. Mo Shahin, an Engineering Consultant retained by the Bureau of Street Services. Testing was performed on trenched street sections as provided by Dr. Shahin. Seventy-eight (78) sections were qualified for the study from a total of one hundred and twenty-two (122) street sections analyzed.

'The Falling Weight Deflectometer (FWD), Geoprobe equipment and a core cutter were used to complete the testing portion for this study. The FWD was used to determine both the deflections of the existing asphalt concrete pavement of the street sections tested and the extent of damagecaused by the utility trench to the surrounding area. The Geoprobe equipment was used to conduct a soil investigation of the subgrade in order to determine soil type and penetrationresistance. A core cutter was used to cut pavement cores to determine asphalt concrete (AC) pavement .thickness of the street sections. •

The AC overlay thickness designs and the total flexible pavement structures were then calculatedfor the patched and non-patched areas.

Included in this report are the testing data and related analysis for the study.

■ iirviin,)v pka"f 'wnla<J mi ci Mr. Ricardo Villacorta of my staff at (213) 485-'0 anvYd

3212MV

ASs?

No eve138x C»&, f?Rt/18

Vw f;«, A/

Ra% If Solomon Duvaatlor (.enroll M-i , ice'- M ;ndai&t L/,

RHSl'RV:

AMML

Table of Contents

Introduction 1

Trench Study Location Map 2

3Procedure

Pavement Deflections (FWD) 5

Existing Pavement Thickness (Coring) 8

Piezocone Penetration Testing 11

Overlay Thickness Designs Required 13

Area of Influence 16

Conclusion 18

Appendices

Appendix A: Overlay Thickness Design.............. ........ .

Total Asphalt Concrete Thickness Design

Appendix B: Falling Weight Deflectometer Test Data...

19

28

37

Determination of Area of Influence 194

Appendix C: Piezocone Penetration Testing Data 235

i

IntroductionAt the request of the Bureau of Street Services of the Department of Public Works,

Standards Division of the Department of General Services has conducted testing research,

overlay design, and analysis for the Los Angeles Street Damage Restoration Fee update

study.

The project was developed and conducted under the guidance of Dr. Mo Shahin, an

engineering consultant retained by the Bureau of Street Services. The study was performed on

Select (high traffic) and Local (low traffic) streets. One-hundred and twenty two possible

sections (each section includes: utility-trenches & respective control areas) were chosen for

testing and from this total, seventy eight sections (thirty local and forty-eight select sections)

were qualified for the study.

The conditions of the qualified streets were essentially surveyed through two methods: A)

Pavement Condition Index (PCI) & B) Pavement Evaluation Testing. PCI (part A) was carried

out by Prof. Shahin’s team and Standards Division conducted all the testing required (part B) for

the study and determined the overlay designs for all the sections analyzed. In addition, attached

are conclusive graphs of the data collected during the update study.

The testing and analysis utilized during the study are as follows:

• The Falling Weight Deflectometer (FWD) was used to determine the deflections

of each trench (patched) and control (non-patched) areas. In addition, the FWD

also determined the extent of damage caused by the utility trench to the

surrounding area.

• Pavement cores were cut to determine the existing pavement structural thickness in both areas.

• Piezocone Penetration Testing (CPTU) was performed to estimate the SPT (N6o)

values, pore pressure and type of soil underneath of each patched and non-patched

area.

• Overlay Thickness designs (DARWin Pavement Design Program) were

performed for all sections including the utility trench and control area.

Included also is the approximate location of all the Local and Select SDRF sites in a City of LA

map. Please see below the location map (Figure 1).

1

&' rt '0~J 'i"x

-■*31P»y. - %

—HBJfi

DAN■■ .*-■

££. J>^ *«hlr45 Vi,...**

v+ »-Bft

%: .*

«?£r *• • /, v J§> # t* •

■, L

|w

»^ j£«* r

$w&

: frX 118i'

.-■’ . : C,

*

$y£zV J-*

*■ yap*&r-J> ■■ <

il/-.*■-. 4

$&c-* ^ <4.as^

>

5a%#-„

LFt "'‘ £r n»ps.:

>1 #■'2ft?

H / ■a‘‘•ffr\

Hi* ■:

ttir. • 'f

i>■% *1\ w

- ' r * ^;AiaCT«i•c.. .•

mi&M

,«

13'4*J J . i .*

V-*, r 14 c Jb • v<w

V. WA'-M .4* . -ft '* *

*t

> ;V ^*wjin'

t'. ,

m >

ill: /• /•*

?**> V * , '■ -i Lid w \ G i t!If ■'tt ,’ * ITsV ’*+, ■ v,*:rf*. -!; T ■■■ < •'>: V'' 18 7.' -

r 90; '

1

1./ -i '

■n. yjjh .

■ w.y ft ;

rH;.J r. 't-

1»«1 * 01

\*'s.

r;

. ^ ;... 1 / *.4T •riii fe'gw:.'- i11 ■=.'i 1

m ‘>

.Vftft-

City of Los Angeles Department of General

SendeesStandards Division

5B^iSLegend

Q = Local Site ■ = Select Site

SDR.F Sites Location Map

*NVi

5gSjAP F

Project Title:STREET DAMAGE RESTORATION FEE

W.O.#S17A0003 March 2017

Figure 1

2

Procedure:Before testing the designated locations, the PCI surveyor determined two areas: A) The

control area and B) The trench area that comprised the utility-trench. Each area was no less than

1500 ft2. The control area did not have any trench inside of the determined area and was at least

10 ft. away from any other trench. The control area also met the following requirements:

a. All selected sections were flexible pavements (PCC pavements are excluded) and

both areas (trench & control) consisted of the same pavement structure, same

thickness, mix, and age.

b. Trench and control areas had the same traffic flow (same lane).

c. The control area was located as close as possible to the trench area and when it was

viable the control area was located immediately adjacent to trench area.

Cores were obtained with the following criteria:

a. Pavement thickness was determined: i) In the trench, ii) Outside the trench, and hi) In

the control area.

b. Cores in each site were located at equal distance from the curb face.

c. The difference in total pavement thickness between the control and outside trench

cores should be less than 1 in.

d. As many cores as necessary (minimum 2 cores) were cut in the “Control FWD testing

area” to verify thickness consistency.

e. The pavement structure was similar in both areas including base.

Falling Weight Deflectometer (FWD) criteria:

a. A minimum of eight deflections were measured on the joint around the trench. The

measurements were obtained by positioning the FWD loading plate so that the edge

of the plate was no more than 0.5 in. away from the joint of the trench. On 95% of

the locations the sensors fell parallel to this joint, on very narrow streets; the sensors

fell perpendicular to the joint. It was noted that trench comers usually showed the

highest deflection.

b. One additional deflection was measured in the center of the trench for trench repair

evaluation purposes.

c. Eight deflections were measured in the control area along the same line as the coring

locations. The spacing between deflections depended on the size of the control area..

3

Deflection readings were taken at equal distance apart on control pavement area that

showed consistent pavement thickness with the tested trench area. If no consistency

of thickness was established, the tested control area was discarded and relocated.

<*>nWsr t

K^y.#

*Irene-,

/

>«« ■ im

Figure 2: Shows both Trench and Control area (S37)

-• ,

.< ^ -

:;

!

Figure 3: Shows FWD to be measured in a control area with consistent Asphalt thickness structure.

4

Pavement Deflections (FWD)A total of 2323 pavement deflection tests were performed during the project. Pavement

deflection is one of the required parameters necessary for pavement evaluation to determine the

pavement structural capacity and to calculate the pavement and overlay design. Deflections,

measured in thousandths of an inch (mils), were directly measured using a track mounted Falling

Weight Deflectometer.

A truck mounted Foundation Mechanics model Jils 20T Falling Weight Deflectometer

(FWD) with an equivalent load of 9,000 pounds was used to measure the pavement surface

deflections of the existing asphalt concrete pavements in the trench (patched) and control (non-

patched) areas. The FWD is a load-deflection device that applies an impulse load by dropping a

mass onto a circular load plate of 6 inches radius placed on the pavement surface to simulate a

moving wheel load. This device uses deflection transducers that measure the resulting pavement

deflections in the “deflection basin.” One transducer is located at the center of the loading plate,

with the remaining six transducers spaced at intervals of 8, 12, 18, 24, 36, and 60 inches from the

center of the plate.

The FWD survey was perfonned by measuring nine (9) pavement deflections in each

trench area (one inside the trench and eight at the outside edge of the trench) and eight (8)

deflections in the corresponding control area. The pavement deflection measurements were

determined in accordance with ASTM Designation: D 4695 - 03.

The measurements obtained are presented in Appendix B. All deflection measurements

were normalized to 9 kips and 68° F using AASHTO Guide for Design of Pavement Structures

1993, AC Temperature Adjustment Factor Table (Figure 5.6, pg. HI-99).

Accumulated deflection, Do (Normalized to 9 Kips and 68° F) of Local Trenches and

respective Controls clearly shows higher accumulated deflection on trenches than controls. The

values are 560 mils and 398 mils, correspondingly. The same observation is seen in Select

Trenches vs. Controls where the values are 504 mils and 334 mils, respectively (See Figure 4).

The average normalized deflection (D0) of Local Trenches was 41% higher than their

corresponding Control. The average normalized deflection (Do) of Select Trenches was also 51%

higher than their corresponding Control (See Figure 5). This shows that the pavement

surrounding each trench has been weakened more than the rest of the pavement section, thereby

accelerating pavement failure under traffic. It was also observed that the damage was higher

among select streets than local streets.5

18.68;

-

11

13.26

11 local, Conttos10.50 ©local Trench

@Select, Control

©Select, Trench

1

I

Accumulated Normalized Deflection, D0 (mils) for Local & Select Sites600

Local, Trench 560.45: *

500faQ Seteec TrenchLocal, Control§ 400 397.89

O*5 300fi1 -ct Control

"\333.72J5 2003E3 ioo<

0

■ Select, Control■ Select, Trench■ Local. Control■ Local. Trench

Figure 4

Average Normalized Deflection, D0 (mils) for Local & SelectSites

oTrenchTrench ControlControl

SelectLocal

Figure 5

6

OO

n

OO

1̂O

QC

O

—r

r-i

Ave

rage

Def

lect

ion,

D„ (m

ils)

D0 (Normalized to 9 kips and 68 F) Distribution Trench vs. Respective Control for Local and Select Sites

#

y- 1.2816X+ 1.6841 R* =0.6321 v*

* Local Site

■ Select Site♦♦

#♦♦♦ ^ 0.8148x + 4.8311

R1 ~ 0,2766■ m■

♦ 4-

4# ♦CM-.

5 10 15 20 25 30 35 40 45Deflection, Control P(, (mils)

Figure 6

7

A regression analysis was performed to determine the relationship between the average

normalized deflections (Do) of Local and Select trenches and their respective controls by means

of a scatter plot (See Figure 6), then a straight line (the best fit of the regression line) that

describes such relationship in the best possible manner was calculated and drawn with the Excel

program.

The trend of the regression line clearly shows that the normalized deflections in the

trench area are higher than the control area or that the trench area has been debilitated by the

utility trench.

oo

§

oo

o

o

oo

ON

’J-1

A-

r-j ciD

efle

ctio

n, T

renc

h D„ (

mils

)

Comparison of erage Pavement I hickness Between Local & Select Sites

12.0

10,6 10.5

10 08,7

8.0

m Local it Select

•5.6..6 0

I f I4.0

2.0

0.0Trench OutTrench In Control

Location

Existing Pavement Thickness (Coring)Existing structural pavement thickness was also determined in order to find out if utility

trenches were properly overlaid to match the original pavement thickness structure and to

calculate the overlay thickness design for both areas. The pavement thickness for each trench

and respective control was determined by coring and are shown in Appendices B & C.

A total of four-hundred-eighteen (418) cores were cut for this entire investigation. Pavement

cores were cut using an Acker Model PT-22 truck-mounted core cutter with an eight-inch core­

bit and Geoprobe Model 6600.

Coring data shows that inside of trench (Trench-In) asphalt concrete thickness is much

less than outside of the trench (Trench-Out) see Figures 7, 8 and 9.

Figure 7

The comparison of the average pavement thickness between Local and Select streets

(Figure 7) demonstrates that in both classifications (Local and Select), the pavement thickness

inside of the trench (Trench-In) was less than in the vicinity of the trench (Trench-Out) and its

respective control. Furthermore, this difference is more significant among Select than Local

streets.

Ave

rage

Thic

knes

s (in

.)

J

4 „♦ ♦ Local Site. 1rench Invs. Control

4

♦♦ ♦♦& Local Site,

Trench Out%'s. Control

4A0.4966s - C 4805 R: - 02I3K♦ A

A

-—-Linear(Local Site.Trench In %'S. Control)

"——Linear(Local Site. Trench Out vs. Control)

A♦♦& ♦♦

♦♦♦♦ ♦4

4

♦I<I 3 4 6 8 9

Control AC Thickness (in.)

Figure 8

The simple linear regression analysis of pavement thickness distribution among inside of

trench (Trench-In), out of the trench (Trench-Out) and respective control area (Figures 8 & 9),

displays basically a relationship of 1 to 1 thickness between Trench-Out and Control area for

both classifications (Local and Select). However the regression analysis between Trench-In and

Control in both classifications are not conclusive due to the inconsistency of the trench repair,

which indicates that utility trenches were not properly restored to match the original pavement

thickness structure.

9

AC Thickness DistributionTrench In & Trench Out vs. Respective Control for Local Sites

10

On

oco

r iTren

ch In

/Out

AC

Thic

knes

s (in

.)

Pavement Thickness Distribution Trench In & Trench Out vs. Respective Control for Select Sites

♦¥'=0.9SS3x+0.4635

R» = 0.8959 ♦ Select Site, Trench invs. Control

X4ta Select Site,

Trench Out vs. Control

♦

M€♦ ¥=.'0.297X4 5.3431

R2 = 0.0847'Linear (Select Site. Trench In vs. Control)

—Linear(Select Site, Trencli Out vs. Control)

*««

*s ♦* +♦ ♦♦ ♦♦ ♦A ♦ ♦♦♦

♦♦♦

%

181614128 106Control Pavement Thickness (in.)

Figure 9

V11

I-

>

<

Figure 10: Existing pavement thickness determination Trench-in & i-out.

10

oo

50

ro

O

O

Ooo

t-vTr

ench

In/O

ut Pa

vem

ent T

hick

ness

(in.)

32•J A

■ Local,Trench In

■ Local. Trench Out

B Local.Control

■ Select, Trench In

■ Select. Trench Out

■ Select. Control

28 28i mum BS■

•-1

■t mmm

IBBControlTrend! Trench Trench Trench Contro

SelectLocal

Piezocone Penetration TestingTwo hundred- thirty- four (234) CPTU Piezocone Penetration Test soundings were

performed and data was collected to a depth of two (2) feet unless refusal depth was reached

(See Appendix C). The CPTU soundings were conducted using a truck-mounted Geoprobe

Model 6600 and a 20 ton capacity cone with a base area of 10 cm2 and a friction sleeve of 150

cm" located above the Piezocone.

Ngo is a parameter classically determined from SPT blow counts and provides an

indication of the relative density and strength of the soil. In this study, N6o is a calculation

resulting from pushing a Piezocone, or cone penetrometer with pore pressure measurement

through the undisturbed soil underneath the pavement and recording data with a computer

program. A comparison average of N60 values for Local and Select trenches and respective

controls are presented in Figure 11.

Comparison of Average N60 for Local and Select Sites

Figure 11

In this comparison, both classifications exhibit the Control N60 values higher than the

Trench-Out which indicates that the utility trench disturbed and debilitated the strength of the

soil under the adjacent pavement. Furthermore, it was determined that 30% of Local and 25% of

Select trenches were treated with liquefied soil cement slurry, consequently improving the

average of the N6o Trench-In values compared with their respective Controls.

11

OC

Oit

en

O

Ave

rage

N40 V

alue

Additional parameters in this dynamic procedure includes measurement of tip resistance

(qc), sleeve friction (fs), and pore water pressure (U2), These measurements determine soil

stratigraphy and corrected SPT energy ratio N6q values. This is all done by operating the

computer programs CPT-log and CPT-pro. On a few occasions, the trench is backfilled with

cemented sand, which is extremely hard. While we are unable to penetrate this layer with the

cone, we were able to penetrate the subgrade below the outside edge of the trench. We found

that N6o values of the subgrade below the outside edge of the trench are lower than the N60 values

of the subgrade in the control. This means that the disturbance of the soil caused by the

excavation of the trench has weakened the surrounding subgrade, which will cause premature

failure of the pavement adjacent the trench.

The Piezocone takes measurements at 2 cm intervals of bearing resistance (qc), unit

sleeve friction resistance (fs), and pore pressure behind cone (U2). All CPTU soundings were

performed in accordance to ASTM D-5778 Standard Test Method for Performing Electronic

Friction Piezocone Penetration Testing of Soils.

All CPTU data was collected by a wireless CPTU cone (serial No. 4130, calibrated on

07/22/2016, traceable to NIST) manufactured by Geotech AB Company. Soil classification is

based on Rf (friction ratio) and qt (corrected cone resistance), Robertson 1986, using CPT-pro

software by Geosoft company.

The measurements that we obtained are presented in Appendix B, with their

corresponding graphical CPTU test results.

Below is a summarized table with percentages of different class!ft.cation of soils

encountered under pavement structures in both groups.

Table 1

Local StreetsClayey- Sensitive

Fine-GrainedClassification Sand Silty-Sand Sandy-Silt TotalSilt

Sites 8 5 4 2 3011

Percentage 27 17 13 7 37 100

Select StreetsSensitive

Fine-GrainedClayey-Classification Sand Silty-Sand Sandy-Silt TotalSilt

Sites 15 14 9 0 4810

Percentage 31 29 19 0 21 100

12

Overlay Thickness Designs RequiredA total of one-hundred-fifty-six (156) overlay thickness designs and (156) Flexible

Pavement designs were calculated. Overlay thickness design is the required addition of

compacted Hot Mix Asphalt (HMA) to an existing pavement in order to sustain predicted

repeated structural loading from traffic over the design life of the pavement.

The overlay and pavement designs were determined utilizing the 1993 AASHTO Guide

for Design of Pavement Structures, AASHTOWare DARWIn 3.1 Pavement Design, Analysis

& Rehabilitation for Windows computer software. The average mid-depth temperature was

calculated based on BELLS3 (Routine testing methods) by using a computer program in which

the computer source code is provided by FHWA.

Accumulated Overlay Thickness Design for Local and Select Sitestty .

201.9£= :m *

Cr

I IMj . • I Irene!;S

XL CYevfiV1:5 - IT ^ '«

S #'SekvF IfeiKr H4»

iSdti!. t.

s 50.9s M5 |HiM£3 5-' 5 .4* i

Jpima■ m

C**!ilto! 1 rrtich ( Valid

Select Sitesfrench

Local Sties

Figure 12

The accumulated overlay thickness design, in inches, for Local and Select trenches and

their respective control areas are presented in Figure 12. In both street classifications, the

accumulated required thickness design is significantly higher in the Trench area than the Control

area (98% higher for Local & 79% higher for Select).

13

Average Overlay Thickness Design Required For Local & Select Sites

4 50

$ 00

* » •

> CIC» ■ B■>

2 50 ■ ■1.702 oo i

anm! 50 ■ 0 HU

! 00■ mmm1

m0 111 jgt■Ml ■■ ■■ ■■II I Ml

I reiidi Control I tench Control

1 .ocal Sites Select Sites

Figure 13

14

In addition, the average overlay thickness design has also been calculated to illustrate how the

weakness inflicted in the pavement by the utility trench has to be restored with “additional

thickness structure” to meet traffic demands when compared with the control average overlay

thickness design required to meet such demands (see Figure 13). Using this figure we conclude

that the Local trench areas require an average of §.84 inches more structure than its respective

control and an average of 1.86 inches more structure is required for the Select trench area than its

respective control. This means that the trench’s surrounding pavement has been negatively

affected by the excavation of the trench, and the effects are very significant and costly in order to

remediate the damage caused by the trench in bringing the structure back to its original capacity.

Ave

rage

Ove

rlay

Thic

knes

s Des

ign (

in.)

New AC Thickness Required vs. Cumulative 18-Kip ESAL Traffic20 Year Performance Life

i

y - 0.2S2xnm8 R!= 0.6493 • Location

PowerTrendline

»

*

5.0 10.0 5.0 20.0

Cuntuiative 18-kip ESAL Traffic (Millions)

Figure 14

15

Asphalt Concrete Thickness Designs versus Traffic: Asphalt concrete thickness

designs were calculated for all trench and control areas (Local and Select collectively) using the

above-mentioned DARWin program and the data acquired in this study were plotted in a graph

as AC Thickness in inches (Y) versus Cumulative 18-Kip ESAL Traffic in Millions (,X). A

power trendline was then calculated and drawn with the Excel program (see Figure 14). The best

fit-line in this case is a power trendline that demonstrates that the AC thickness designs are

directly related to traffic (more traffic = more thickness) notwithstanding the reduction of the

positive rate of thickness/traffic at high levels of traffic. Therefore the cost to repair the damage

in Select or high traffic pavement is a lot higher than the cost in Local or low traffic pavement.

O- ^

OO O

n OO A 50 if,

*T M

l-J Isj

NJ

t'O l-J

KJ U>

4-■t—

. '_/i Os J

OO

v© o

OL-

J LA

C T

hick

ness

(in)

Area of InfluenceA few trenches were also selected for Area of Influence determination. Starting at the

edge of each trench, deflections were measured one foot apart, moving away from the edge until

the change in deflection from the previous deflection reached near-zero. The purpose of this

testing was to determine the distance away from each trench where the subgrade was found to be

unaffected by the utility cut. The area from the edge of the trench to where change in deflection

is near-zero is called the Area of Influence. It was determined that the area of influence

fluctuates between 8 to no more than 10 feet. (See figures 15, 16,17, and 18)

1■«!■

Vt v, :m m ■■ naSSI W

V - fjmifjfi i-T_

,inik

mm

I

Defections

Figure 15: Area of Influence Select Trench No. 74

16

:Distance from Trench (ft.)

Figure 16 Determinations of “Area of Influence" for Select Trenches No. 8 & 37

Area of Influence7/

/ 10’10’-X

m.

i/ At|*haK cim-iiiSub-grade J

Trench

Figure 17: Sketch of Area of Influence under pavement.

17

Deflection Do (Normalized to 9 kips & 68°F) vs. Distancefrom Trench

PavementTrench

l?

O

*

!>

#

/if*

fd

/

/

Oo

oOf;

Nor

mal

ized

Defle

ctio

n I),,

(mils

)

Overlapping Area of Influence

10’ 10’L > LL 77 7 7

| Imi

/I■iAsphalt Pavement

Sub gradeL JNew Trench

Old Trench

Figure 18: Sketch of Overlapping Area of Influence under pavement.

Figures 17 & 18 show that regardless of the life of asphalt concrete pavement, utility cuts

weaken the underlying subgrade of the pavement thus requiring a thicker overlay.

ConclusionBased on all the tests and analysis performed in this study, it is evident that there is

significant damage inflicted by utility trenches to the adjacent pavement structures and

underlying subgrade. Regardless of the age of the asphalt concrete pavement, the damage to the

underlying subgrade of the pavement adjacent to the utility cuts remains significant;

consequently, the overlay thickness design to repair such damage practically doubled the overlay

thickness design required on the non-patched area as intended for future traffic in the same

section.

Furthermore, the study indicated that utility trenches were not properly restored to match

the original pavement condition. It is also evident that the damage and the repairs on higher

traffic streets (Select) due to utility trenches are higher and therefore, costlier than low traffic

streets (Local).

Appendices