Use of Shredded Rubber in Unbound Granular Flexible ... · Use of Shredded Rubber in Unbound...

35
Use of Shredded Rubber in Unbound Granular Flexible Pavement Layers Paper Prepared for Presentation at the 1995 Transponation Research Board Meeting by Richard H. Speir Graduate Research Assistant Ph: (301) 405-1249 Fax: (301) 405-2585 and Matthew W. Witczak Professor Ph: (301) 405-1941 Fax: (301) 405-2585 Civil Engineering Department University of Maryland College Park, MD 20742 1 August 1994

Transcript of Use of Shredded Rubber in Unbound Granular Flexible ... · Use of Shredded Rubber in Unbound...

Use of Shredded Rubber in Unbound Granular Flexible Pavement Layers

Paper Prepared for Presentation at the 1995 Transponation Research Board Meeting

by

Richard H. Speir Graduate Research Assistant

Ph: (301) 405-1249 Fax: (301) 405-2585

and

Matthew W. Witczak Professor

Ph: (301) 405-1941 Fax: (301) 405-2585

Civil Engineering Department University of Maryland

College Park, MD 20742

1 August 1994

Speir Witczak

ABSTRACT

2

The major objective of this research was to conduct a feasibility study into the use

of shredded rubber as a partial replacement for aggregate within conventional base and

subbase materials within a flexible pavement system. A graded aggregate base and sand

subbase meeting specifications for the Maryland S tate Highway Administration were used.

The rubber used in this study consisted of a 3rd-4th stage shred product with 60 -

70% retained on a 9.5 mm (3/8 in) sieve. This size was selected because of its relatively

inexpensive cost to produce and because of its adaptability to an aggregate blend.

Modified and Standard Proctor tests, CBR tests and Mr tests were conducted on

. the base/subbase-rubber blends with up to 15% rubber content by weight. The aggregate

base blend resulted in significant decreases in both CBR and non-linear Resilient Modulus

at 15% rubber. These significant reductions led the authors to conclude that the use of

shredded rubber in a dense-graded aggregate base course is not feasible.

In contrast to the granular base, the sand subbase-blends resulted in very

insignificant changes to the CBR, friction angle, permeability and Mr at higher rubber

percentages. It was concluded that the sand-rubber subbase exhibits little, if any, change

compared to the virgin sand subbase material. As a result, the use of shredded rubber may

be a technically feasible alternative in the construction process.

Finally, two constitutive models were used in the Mr analysis; the Conventional

Kl, K2 Model and a Universal Model incorporating an octahedral stress, 'oct, term (kl,

k2, k3 model). Direct comparisons revealed greatly improved predictability and accuracy

with the Universal Model to assess the non-linear behavior ofboth aggregate types

evaluated.

Speir Witczak

INTRODUCTION

3

Section 1038 of the 1992 Intennodal Surface Transportation Efficiency Act

(ISTEA), "Use of Recycled Paving Material," established that each state shall incorporate

a minimum utilization requirement for the use of recycled rubber in asphalt pavement.

While the implementation of this regulation is still under discussion, investigative studies

regarding the potential benefit of rubber in asphalt mixtures has intensified in recent years.

Currently there are several major alternatives and numerous ongoing studies to

implement and achieve the ISTEA requirements. One process combines crumb rubber

into the asphalt cement fonning an asphalt-rubber blend. This procedure is commonly

referred to as the "wet process." Another procedure incorporates crumb rubber as a

partial aggregate substitute in the mix. This process is known as the "dry process."

An alternative, potential process for the disposal of waste tires in pavement

systems incorporates a less finely ground scrap tire particle, hereafter called shredded

rubber, into unbound aggregate base and subbase material. This non-asphalt alternative

possesses several potential advantages over the wet and dry asphalt processes because the

shredded rubber need not be as finely ground as the crumb rubber. Shredded rubber is

only a third- or fourth-stage shredding as opposed to a sixth- or seventh-stage crumb

necessary for the wet and dry processes. Much of the cost associated with incorporating

crumb rubber into an asphalt mix is a result of the refinement of the rubber panicles. Cost

to produce the crumb for the wet and dry processes may range between $0.10 to $0.30

per lb versus the cost to produce shreds which typically range in price between $0.01 and

$0.03 per Ib (1).

The potential for the use of the shreds in base or subbase materials cannot be

understated. Given the requirements to utilize scrap tire in 20% of the tons of asphalt mix

used by 1997, the utilization of shreds in the granular layers could save states between

$0.07 and $0.29 per lb of rubber used. In addition, the potential quantities of shreds

Speir Witczak

utilized may increase due 10 significantly larger quantities of base and subbase material

used in a new construction project as well as a greater percentage of shredded rubber used

per unit weight of the base/subbase material.

The use of shredded rubber as an aggregate substitute is a relatively new concept.

Several states have conducted or are conducting tests on various scrap tire applications.

Oregon, Vermont, and Minnesota, to a name only a few, currently have field studies

underway which utilize the shredded rubber as a lightweight fill.

Although the results of the field work are promising, these projects still fall short

of studying the feasibility of adding the shredded rubber as an aggregate substitute for

conventional unbound base/subbase layers of a pavement system. In most cases, the

. rubber was placed in a separate layer in the structure without trying to integrate it into one

of the existing layers. Based on the lack of information for rubber in unbound granular

material, this study focused on trying to characterize those types of aggregate-rubber

blends.

STUDY OBJECTIVE

Given the lack of technical data available in the use of shredded rubber as an

aggregate substitute in base and subbase materials, this research was conducted from the

perspective of a preliminary feasibilitv studv. The object of this study was to investigate

the possibility of using shredded rubber as a partial replacement for aggregate within

conventional unbound aggregate base and subbase material within a pavement system. To

meet this objective, typical unbound base and subbase materials utilized by the Maryland

State Highway Administration were obtained and then blended with the shredded rubber.

These blends were then studied to determine the effects of the rubber on the strength and

dynamic response of each aggregate. Two different types of aggregate were selected: a

graded aggregate base (GAB) and a sand subbase material.

Speir Witczak

LABORATORY TESTING

Material Characterization Tests

j

In order to provide the necessary reference data for all the materials used, certain

standard material characterization tests were conducted initially. Sieve analysis, specific

gravity, and absorption tests were performed on all materials. In addition, Atterberg limits

tests were conducted on aggregate samples only. Figure 1 and Table 1 are the results of

these preliminary tests.

Strength Characterization Tests

To investigate the strength characteristics of the aggregate-rubber blends,

moisture-density compaction tests and California Bearing Ratio (CBR) tests were

conducted. In developing the blends, percentages of rubber shred were defined on a total

weight basis. Because no previous knowledge existed concerning typical rubber

percentages to be used, the upper boundary percentage was arbitrarily selected as 15

percent. An intermediate level of 7.5 percent was established as well as a zero percent

control. The selected percentage of shredded rubber (0-15%) was found to be a good

range for characterizing the strength effects of the blended material.

For the purposes of establishing the moisture-density relationship of the aggregates

as well as the blends; two levels of compactive energies were employed: a modified

energy equal to 2694 kJ/m3 (56,250 ft-Ib/ft3) and a standard energy equal to 593 kJ/m3

(12,375 ft-Ib/ft3). Molds with a diameter of 15.2 em (6.0 in) and having a volume of

0.0021 m3 (0.075 ft3) were used to compact all specimens. This mold size was used to

facilitate tests of the CBR strength on the compacted specimens. The specimens were

compacted by a mechanically operated metal rammer which was equipped with a device to

control the height of drop of the rammer and which uniformly distributed the drops around

the specimen surface.

The number of specimens required to characterize the optimum dry density along

with the optimum moisture content varied depending on the material being used .

...

Speir Witczak

6

Typically, the GAB blend was more difficult to characterize than the sand-rubber blend.

As a minimum, however, no fewer than four points were used to characterize the

moisture-density relationship. Table 2 provides a summary of the optimum dry density

with the corresponding optimum moisture content. This table reveals the expected

decrease in dry density with increasing rubber content for both the GAB and sand. Figure

2 is a graphical representation of the effects of the rubber content on Y dopt. The Glopt

determined by each Ydopt are also plotted against the rubber content in Figure 3. A rise in

Glopt in the GAB-rubber blends is more than likely a result of the rubber shreds retaining

more water than the aggregate they replaced. The sand, on the other hand, reacts in an

inverse relationship between the moisture content and the rubber percentage. Because of

. the introduction oflarger particles into the otherwise fine sand material, many of the grain-

to-grain contacts of the sand particles are displaced. The fine sand material then takes on

characteristics of an increasingly coarse material as more rubber is introduced.

Two specimens were tested using the AASHTO T 193 CBR test method at each

moisture content, each rubber content, and both compaction conditions. One specimen

was tested in the as-molded condition, and the second specimen was immersed in a

soaking tank for a 96-hour soak prior to testing. No swell tests were performed on the

soaked specimens as the blended materials were considered free draining, and swelling

was assumed to be insignificant. All CBR values were plotted and a maximum value

corresponding to the Glopt was recorded for each rubber content and each compactive

energy. A summary of the CBR corresponding to the optimum moisture content is

provided in Table 3. In addition, Figure 4 shows the effects of the rubber percentage on

CBR. Analysis of this graph reveals that the GAB-rubber blends result in a significant loss

in strength for the as-molded, modified compaction conditions. At 0% rubber, a CBR of

93 is measured at the optimum moisture content. At 15 % rubber, however, the CBR

value drops to 13. The GAB-rubber blend CBR values at standard compaction and

Speir Witczak

7

saturated conditions have similar trends indicating a major loss in strength for all

conditions.

The sand, on the other hand, results in a much more positive response with the

addition of the rubber. Given the naturally occurring variation in typical CBR results, it

can easily be argued that for the range of rubber percentages investigated, there are no

changes in the CBR values from 0% to 15% rubber. The nearly horizontal lines in Figure

4 suggest that adding up to 15% rubber to the sand has no adverse affect on the CBR

strength of the material.

Sand Triaxial Compression Tests

In order to verifY the CBR-rubber trends found for the sand-rubber blends, two

. additional tests were carried out on this material: the triaxial compression test and the

constant head permeability test. Table 4 summarizes the results of the triaxial

compression test. The results shown in this table indicate that the angle of internal friction

is almost independent of the rubber percentage used. This conclusion is identical to CBR­

rubber trends found on the sand subbase and obviously lend support to the conclusion that

the shear strength (as measured by both CBR and friction angle) are independent of the

rubber percentage used.

Sand Constant Head Permeability Tests

Constant head permeability tests were also conducted on only the sand-rubber

specimens. The coefficients of permeability determined from these tests are shown in

Figure 5. These results clearly indicate that the effect of increasing rubber percentage on

the coefficient of permeability is somewhat insignificant. A slightly enhanced drainability

condition does appear to occur at the high (15%) rubber content.

Resilient Modulus Tests

In general, AASlITO T 294-92, Test for Resilient Modulus of Unbound Granular

Base/Subbase Material and Sub grade Soils, was utilized in cpnducting the Mr test for the

GAB and sand-rubber blends. A repeated axial load sequence consisting of 0.1 second

Speir Witczak

8

haversine load application followed by a 0.9 second dwell period was the basic load pulse

applied during the test. This load sequence was repeated for a predetennined number of

applications. In addition to the vertical dynamic loading, the specimen was simultaneously

subjected to a static confining pressure. The whole procedure was repeated at various

confining pressures and load levels until a representative number of sequences were

performed to sufficiently characterize the material. For this phase of the procedure, the

AASHTO method was slightly modified to incorporate additional stress sequences. This

was done in order to allow for more stress conditions (data points) with which to evaluate

the Mr results. Table 5 identifies the stress sequence used in the non-linear Mr evaluation

of all materials.

The most common model used for characterizing the non-linear behavior of

cohesionless material was originally developed by Hicks and Monisrnith (;;'). This model is

referred to as the Conventional Model in this paper. It is expressed by:

,'vI, = Kl x gA"

where:

e = Bulk stress = CJI + CJ2 + CJ3

Kl, K2 = material regression constants

(1)

In the last decade, many researchers, such as May and Witczak (J.); Brown and Pappin (1);

and Uzan et. al. (2), have questioned this model for characterizing all cohesionless

material. Consequently, Uzan et. al. (2.) have proposed a "Universal Model" equation

which incorporates a deviatoric stress component into the Conventional Model. This

relationship is:

where:

'oct =

kl, k2, k3 =

octahedral stress = fi (Jd ; for the triaxial test case 3

material regression constants

(2)

Speir Witczak

The Mr results from this study were evaluated using both models.

In looking at the effects of rubber on Mr for the conventional analysis, it is

important to note the change in K 1 and K2 values with respect to increasing rubber.

9

Table 6 and Figures 6 and 7 show the trend in the effects of the rubber percentage upon

the Kl, K2 values for the GAB. As rubber is added, a significant decrease in the KI value

is noted for all conditions. Conversely, values ofK2 increase for all GAB-rubber

conditions. Both conditions are examples of an increased level of non-linear response of

the aggregate as rubber is added.

A review of the sand KI values (Figure 6) shows that although the general trend is

similar to the GAB-rubber blend, the decrease is far less drastic. In this plot, the rubber

·clearly has a greater effect on the Kl values of the GAB-rubber blends than the Kl values

of the sand-rubber blend. Further study of Figure 7 shows somewhat unexplainable trends

for the sand-rubber K2 values. The saturated sand-rubber specimens behave in a manner

similar to the GAB-rubber blends as they increase with increasing rubber percentages.

Quite the opposite is true, however, for the as-molded specimens. In this case, the K2

values appear to decrease. The obvious disparity in trends of the curves suggests that

some other factor is responsible for influencing the sand-rubber specimens besides just the

amount of rubber being added.

The introduction of-toct with the Universal Model was found to have a profound

effect on the overall trends for the regression constants k I, k2, and k3 as well as the

ultimate model accuracy. Table 7 summarizes these values for both the GAB and sand­

rubber mixes. In addition, Figures 8, 9, and 10 show a graphical representation of these

results. In the case of k I in Figure 8, there is a decrease in the value as the rubber is

increased in both materials. This finding is identical to the case of the K I value of the

conventional analysis. As before, the decrease in the GAB-rubber kl is significantly more

than the decrease in the sand-rubber kl. Figure 9 clearly shows an increase in the k2

values for both materials. In this analysis of the Universal Model, there is no irrational

...

Speir Witczak

behavior of the k2 values. Figure 10 indicates that the k3 values become more negative

indicating a more significant impact of the 'oct on the Mr results.

The initial analysis of the goodness of fit statistics for the individual regression

10

models used with the conventional (KI,K2) approach provided the first indication of the

accuracy limitations associated with this model. As Table 8 shows, for the case of the

GAB-rubber blends, the range in R2 values fell between 0.870 and 0.977 with an average

R2 0fO.946. These values indicate good predictability and accuracy between the model

and the measured values for the GAB-rubber specimens. However, a similar review of the

KI, K2 model for the sand-rubber specimens is indicative of a much poorer agreement

between predicted and observed results. For the sand-rubber tests, R2 values range from

. 0.643 to 0.965 with a mean R2 value of 0.826. These R2 values represent a noticeable

decline in the agreement between the actual versus predicted values of Mr for

the Conventional Model. The wider range ofR 2 values may also present a possible

explanation for why the K2 values of other sand-rubber specimens show inconsistent

trends.

A review of the R2 values for the Universal Model analysis in Table 8 clearly

shows the improved accuracy obtained by this model form. For the GAB-rubber blends,

the RZ values range from 0.948 to 0.996 with a mean value of 0.987. Furthermore, the

sand-rubber blend R2 values range from 0.887 to 0.992 with an average of 0.945. It is

obvious that the higher R 2 values will produce better predicted results and constitute a

better model compared to the Conventional Model.

To further highlight the differences between these two models, Figures 11 and 12

are plots of the observed versus predicted Mr values. Figure 11 represents a plot of all

GAB and sand values using the Conventional Model analysis, while Figure 12 represents

an analysis using the Universal Model to determine the predicted values. It is obvious

from the comparison of these two figures and the regression statistics shown on each plot

Speir Witczak

that the Universal Model (kl, k2, k3) results in a greatly improved and more accurate

model form to characterize the non-linear Mr of the materials investigated in this study.

CONCLUSIONS

Shredded Rubber in GAB

11

Addition of shredded rubber causes a decrease in the 'Y dopt for both modified and

standard compaction

(Dopt increased with increasing rubber content

Addition of 15% shredded rubber caused significant reductions in CBR values

Addition of as little as 5% shredded rubber caused moderate reductions in Mr

values

. Based on these observations, this study concluded that the use of shredded rubber in a

dense-graded aggregate base course is not highly feasible.

Shredded Rubber in Sand

Addition of shredded rubber resulted in a decrease of the Y dopt for modified and

standard compaction

(Dopt decreased as rubber content increased

Increasing rubber percentages had almost no affect on CBR values

The introduction of rubber produced little change in the angle of internal friction

The coefficient of permeability increased slightly as rubber was added

Increasing rubber percentages caused slight reductions in observed Mr values

Because the observed properties of the sand in several cases were unaffected by the

addition of the rubber, this study concluded that the use of shredded rubber in sand

subbases may be a technically feasible alternative to the use of rubber in pavement systems

and that further research is warranted.

Constitutive Model Comparison

The Conventional Model displayed limitations in accuracy in the analysis of the

measured values of the sand-rubber blends

Speir Witczak

Analysis of the sand-rubber specimens using the Universal Model produced an

increase in the mean R2 values for the individual analysis of 0.945 as compared

with a mean R2 of 0.826 from the Conventional Model

12

Direct comparisons of the regression statistics from the combined observed versus

predicted plots revealed higher R2 and lower Se/Sy values for the Universal Model

The Universal Model analysis produced better accuracy and predictability than the

Conventional Model analysis for the data obtained in this study. In addition, the extra data

collected by the additional testing sequences in the University of Maryland procedure was

extremely helpful in this comparative analysis. Based upon these results, it is highly

recommended that the Universal Model be used to analyze Mr results and that the testing

sequence shown in Table 5 be considered by AASHTO for Mr evaluation of cohesionless

materials.

RECOMMENDATIONS FOR FUTURE WORK

Field Studies

Further studies should be undertaken to better understand and expand the

implication of the laboratory results of the sand-rubber blends. In addition to laboratory

work, field studies (demonstration projects) are needed to determine the best way to

incorporate and mix the shredded rubber into the sand subbase. Some difficulties may

arise, for instance, with in-place mixing to obtain uniform blending. This may result in the

inability to obtain adequate in-place densities. Once completed, these test sections should

be observed for long term performance monitoring associated with the aggregate-rubber

layers.

Laboratory Research

Although the study has examined the sand-rubber mix as a subbase material, there

are additional related areas which should be addressed in the laboratory. Repeated load

permanent deformation behavior of the rubber blends should be evaluated in the

laboratory. In addition, environmental concerns warrant further laboratory tests that

Speir Witczak

13

evaluate possible harmfulleachates from the material in its blended state. Another area of

interest is the reaction of rubber in the freezelthaw conditions in the aggregate layer.

ACKNOWLEDGMENTS

The authors are grateful to the technical and financial assistance provided by the

Maryland Department of Transportation (SHA) for completion of the project study

described in this paper. The views and opinions presented are those of the authors and not

theMDOT.

REFERENCES

1. Witczak, M.W. Use of Ground Rubber in Hot Mix Asphalt. State of the Art Synthesis

. Report. State Highway Administration, Maryland Department of Transportation, Jun 91.

2. Hicks, R. G. and C. 1. Monismith "Prediction of the Resilient Response of Pavements

Containing Granular Layers Using Non-Linear Elastic Theory. Proceedings,3rd

International Conference on the Structural Design of Asphalt Pavements, London,

England, S ep 76.

3. May, R.W. and M. W. Witczak "Effective Granular Modulus to Model Pavement

Responses." TRR 810, TRB, National Research Council, Washington, D.C., 1981,

pp. 1-9.

4. Brown, S.F. and J. W. Pappin "Analysis of Pavements with Granular Bases." TRR 810.

TRB, National Research Council, Washington, D.C., 1981. pp 17-23.

5. Uzan, J., MW. Witczak, T. Scullion, and R.L. Lytton, "Development of Validation of

Realistic Pavement Response Models," Proceedings, 7th International Conference on

Asphalt Pavements, Nottingham, England, 1992

Speir Witczak

LIST OF TABLES

Table 1

Table 2

Table 3

Table 4

Table 5

Table 6

Table 7

Table 8

Specific Gravity Results

Summary of Optimum Dry Densities and Optimum Moisture Contents

Summary ofCBR Values at Optimum Moisture Contents

Sand Triaxial Compression Test Summary

University of Maryland Testing Sequence

Summary of Conventional Model Constants

A. Summary of Average Kl Values

B. Summary of Average K2 Values

Summary of Universal Model Constants

A. Summary of Average kl Values

B. Summary of Average k2 Values

C. Summary of Average k3 Values

Mr Regression Summary

Speir Witczak

TABLE 1 Specific Gravity Results GAB

Bulk S.G. 2.795

S.S.O. S.G. 2.807 Coarse Aggregate Apparent S.G. 2.827

% Absorption 0.399

Bulk S.G. 2.762

S.S.O. S.G. 2.764 Fine Aggregate Apparent S.G. 2.768

% Absorption 0.078

Mineral Apparent S.G. 2.884 Filler

Blend Apparent S.G. 2.809

... + Note - not tested

Sand Rubber ••• 1.082

••• 1.140

••• 1.148

••• 5.350

2.620 1.169

2.625 1.227

2.632 1.241

0.169 4.950

2.857 •••

2.639 1.157

Speir Witczak

-._- --_ .. _----

, TABLE 2 Summary of Optimum Dry Densities and Optimum Moisture Contents 0% rubber 7.5% rubber 15% rubber

Moisture Dry Moisture Dry Moisture Dry Content Density Content Density Content Density

(%) (lbs/ft" 3) • (%) (Ibs/ft" 3) • (%) (lbs/fC3) * Modified 5.0 147.0 6.2 135.1 8.4 119.0

GA8 Compaction Standard 5.8 145.0 6.1 129.7 8.9 120.6

Compaction

Modified 14.0 106.6 13.0 101.5 11.0 97.6 Sand Compaction

Standard 14.0 106.4 11.8 100.8 11.0 97.0

- --- -...f()mpaction

* Note - 1 Ib/ft"3 = 0.157 kN/m"3

'1, Speir Witczak

,

r .·;;:idiF:·f· i:::i"" "

TABLE 3 Summary of CBR Values at Optimum Moisture Contents 0% rubber

Moisture Moisture Content As-molded Soaked Content

1%1 1%1 1%1 1%1 Modified 5.0 93 79 6.2

GAB Compaction Standard 5.8 55 52 6.1

Compaction Modified 14.0 21 14 13.0

Sand Compaction Standard 14.0 15 11 11.8

Compaction -

7.5% rubber 15% rubber Moisture

As·molded Soaked Content As-molded Soaked 1%1 (%1 (%1 1%1 1%1 48 41 8.4 13 13

26 25 8.9 14 14

19 14 11.0 21 16

16 11 11.0 17 15

,

Speir Witczak

,,..... .... 5.0 ... ""T ......... u , ...... 10. ........... "" ........................ __ ....................

Specimen #1 Dry

Rubber Sigma-l f Sigma-3f Density (%) (psi)' (psi)' (pet)· •

0 23.92 6.00 107.0 Modified

Compaction 7.5 25.22 5.86 102.3

15 26.41 5.74 98.3

0 24.42 5.96 107.0 Standard

Compaction 7.5 23.04 5.86 100.9

15 27.32 5.94 97.5

, Note - 1 Iblin - 2 = 6.89 kNlm - 2 .. Note-llblft"3 = 0.157 kNlm"3

Specimen 112

Sigma-l f Sigma-3f (psi)' (psi)'

57.71 13.35

56.22 13.23

56.77 13.13

55.01 13.27

53.72 13.18

55.27 13.20

Specimen #3 Dry Dry Friction

Density Sigma-l f Sigma-3f Density Cohesion Angle (pcO" (psi)' (psi)' (pc!)' • (psi)' (degrees)

106.8 82.86 19.58 106.8 0.00 37.9

102.3 88.21 20.51 102.3 0.00 38.4

98.5 87.08 20.48 98.5 0.00 39.0

1 07.0 84.26 20.60 107.0 0.00 37.5 ,

100.8 81.97 20.52 100.2 0.00 36.9 ,

97.5 85.45 20.44 97.0 0.00 38.6 I

Speir Witczak

TABLE 5 University of Maaiand Testing Seguence Sequence ac (psi)a ad (psi)a Nreps

Ob 10 10 500 1 3. 1.5 100 2c 3 , 100 j

,c , 6 100 j j

4c , 9 100 j

5 5 2.5 100 6c 5 5 100 7c 5 10 100 8c 5 15 100 9 7.5 3.75 100 10 7.5 7.5 100 11 7.5 15 100 12 7.5 22.5 100 13 10 5 100 14c 10 10 100 15c 10 20 100 16c 10 30 100 17 15 7.5 100 18c 15 15 100 19c 15 30 100 20 15 45 100 21 20 10 100 22c 20 20 100 23 c 20 40 100 24 20 60 100

a_I Ib/in2 = 6.89 kN/m2

b _ Preconditioning sequence c _ Represents sequence also used in AASHTO T294

Speir Witczak

TABLE 6 Summary of Conventional Model Constants

A Summary of Average K1 Values Rubber Modified Compaction Standard Compaction

(%) As-molded Saturated As-molded Saturated 0 6560 7150 7140 6216

GAB 2.5 4608 5048 6149 4368 5 4202 2993 3211 2719

0 3459 3597 3872 3370 Sand 7.5 3064 2731 2424 1810

15 2389 1560 2342 1741 Note: K 1 values shown in psi (x 6.89 = kPa)

B Summary of Average K2 Values Rubber Modified Compaction Standard Compaction

(%) As-molded Saturated As-molded Saturated

0 0.482 0.471 0.459 0.492 GAB 2.5 0.534 0.517 0.466 0.535

5 0.528 0.596 0.573 0.602

0 0.527 0.509 0.496 0.539 Sand 7.5 0.465 0.479 0.507 0.571

15 0.417 0.523 0.430 0.558

Speir Witczak

TABLE 7 Summary of Universal Model Constants

A Summary of Average k1 Values Rubber Modified Compaction Standard Compaction

1%) As-molded Saturated As-molded Saturated 0 4896 5326 5279 4566

GAB 2.5 2980 3437 3846 2739 5 2468 1663 1934 1455

0 2266 2337 2628 2195 Sand 7.5 1582 .. 1149 1072 710

15 880 473 786 486 Note: Kl values are shown In pSI Ix 6.89 = kPa)

8 Summary of Average k2 Values Rubber Modified Compaction Standard Compaction

(%) As-molded Saturated As-molded Saturated

0 0.611 0.603 0.590 0.624 GAB 2.5 0.723 0.683 0.672 0.741

5 0.761 0.853 0.800 0.868

0 0.712 0.704 0.669 0.725 Sand 7.5 0.760 0.865 0.860 0.976

15 0.848 1.028 0.892 1.009

C Summary of Average k3 Values Rubber Modified Compaction Standard Compaction

1%) As-molded Saturated As-molded Saurated

0 -0.109 -0.114 -0.106 -0.101 GAB 2.5 -0.154 -0.136 -0.170 -0.171

5 -0.191 -0.212 -0.192 -0.214

0 -0.154 -0.164 -0.146 -0.152 Sand 7.5 -0.250 -0.318 -0.287 -0.328

15 -0.351 -0.402 -0.368 -0.431

Speir Witclak

'I T ~BlE 8 Mr Regression Summary

, A. Conventionsl Model

Trial 1 Trial 2 Trial 3 Avorages Aubber % R2 So S, SefSy R2 S, S, 5ef5y R2 So S, 5ef5y R2 So S, SufSv

0 0.970 0.0218 10931 1.99E-06 0.972 0.0231 11951 1.93E-06 0.977 0.0224 12966 1.73E·06 0.973 0.0224 11949 1.88E-06 As·molded 2.S 0.965 0.0334 12499 2.67E·06 0.932 0.0348 10456 3.33E·06 0.952 0.0345 11657 2.96E·06 0.950 0.0342 11537 2.99E·06

Modlfiod S 0.933 0.0421 10242 4.11E-06 0.922 0.0418 11392 3.67E·06 0.928 0.0420 10817 3.89E·06 Compectlon 0 0.963 0.0270 12862 2.10E·06 0.964 0.0231 11561 2.00E-06 0.957 0.0286 13316 2.15E-OS 0.961 0.0262 12580 2.08E-OB

Soaked 2.5 0.956 0.0314 11792 2.S6E·06 0.951 0.0322 11825 2.72E-OB 0.954 0.0318 11809 2.69E·06 GAB 5 0.934 0.0453 110B6 4.09E·OB 0.928 0.0471 11177 4.21E-06 0.931 0.0462 11132 4.15E-06

0 0.958 0.0270 12054 2.24E·06 0.970 0.0227 12154 1.87E·Q6 0.964 0.0249 12104 2.05E·06 As·molded 2.5 0.870 0.0438 9890 4.43E·06 0.928 0.0381 11017 3A6E·OS 0.941 0.0379 119BO 3.16E·OS 0.913 0.0399 10962 3.S8E·06

Slimdaru 5 0.926 0.0463 10536 4.39E·06 0.9S4 0.0369 105Bl 3.49E·06 0.940 0.0416 ,0559 3.94E-06 Compaction 0 0.968 0.0229 11228 2.04E·06 0.961 0.0275 12415 2.22E·06 0.965 0.0252 11822 2.13E-OS

Soaked 2.5 0.939 0.0387 11414 3.39E-06 0.947 0.0365 11794 3.09E·06 0.943 0.0376 11604 3.24E·OS 5 0.932 0.0468 10886 4.30E-06 0.922 0.0476 10242 4.6SE·06 0.927 0.0472 10564 4.47E-06 0 0.940 0.0340 8855 3.84E·06 0.940 0.0304 8929 3.40E-06 0.965 0.0344 7959 4.32E-06 0.948 0.0329 8581 3.86E-06

As·molded 7.S 0.838 0.0614 6118 1.00E·OS 0.784 0.OS87 50B8 1.1SE·05 0.B44 0.0563 5433 1.04E-05 0.822 0.0588 5546 1.06E·05 Modified 15 0.650 0.082S 3601 2.29E·OS 0.680 0.0813 3827 2.12E-05 0.665 0.0820 3714 2.21E·05

CompactIon 0 0.949 0.0380 8613 4.41E-06 0.942 0.0339 7678 4A2E·06 0.946 0.0360 8146 4.41 E·06 Soaked 7.S 0.799 0.0697 5BO, 1.20E·OS 0.763 0.0796 6219 1.28E·05 0.781 0.0747 6014 1.24E·OS

Sand 15 0.706 0.0961 4459 2.16E-05 0.758 0.0878 42S4 2.0SE·05 0.732 0.0920 4362 2.11E·OS 0 0.964 0.0304 S644 3.52E-06 0.935 0.0359 8097 4.43E·06 0.944 0.0314 7633 4.11E-06 0.948 0.0326 8125 4.02E·06

As-molded 7.5 0.82B 0.0669 5995 1.12E-05 0.822 0.0665 6007 1.11E·05 0.825 0.0667 6001 1,11E-05 Standard 15 0.687 0.OB34 3947 2.11E-05 0.665 0.0864 405S 2.13E-05 0.676 0.0949 '001 2.12E·05

Compaction 0 0.949 0.0378 9629 3.93E-OS 0.929 0.0403 8951 4.50E·06 0.939 0.0391 9290 4.21E·06 Soaked 7.5 0.847 0.0704 6555 1.07E·05 0.835 0.0745 6281 1.l9E-05 0.841 0.0725 6418 1.13E·05

15 0.683 0.0942 3837 2A6E-OS 0.643 0.0998 3900 2.56E·05 0.663 0.0970 3869 2.51E·05

B_ Universal Model

Sood

Standard Compaction

Soaked

As·moldod

1 Rubbor % I R: Trial 1

So t...§v ~ISy R2

1.64E·01

1.0198 I ,VUOU '.ULe-uu u.::>o" u.u'<uo IV........ ....v ..... · ......

E-06 0.992 0.0147 8929 1.65E·06 0.985 0.0175 7959 2.20E-06 0.984 6.0194 8581 2.26E-06 ~ 06 0.914 0.0371 5088 7.29E-06 0.944 0.0338 5433 B.22E-06 0.936 0.0349 5546 6.35E-06

Modiliod 15 0.B87 0.0470 3601 1.31E·05 0.893 0.0470 3827 1. 23E-05 0.890 0.0470 3714 1.27E·05 Compaction 0 0.987 0.0193 8613 2.24E·OS 0.987 0.0162 7678 2.11E-06 0.985 0.987 0.0178 8146 2.18E-06

Soaked 7.5 0.931 0.0407 5809 7.01E-06 0.912 0.0485 6219 7.80E·06 0.922 0.0446 6014 7.4E·06 15 0.919 0.0503 4459 t.13E-OS 0.924 0.0493 4264 1.16E-05 0.922 0.0498 4362 1. 14E-05 o 0.966 0.0176 8644 2.04E·06 0.986 0.0761 8097 9.40E-06 0.984 0.0169 7633 2.21E-06 0.986 0.0369 8125 4.S5e-OS

All-molded 7.5 0_938 0.0412 5995 6.87E-06 0.943 0.0378 6007 6.29E-06 0_941 0.0395 6001 6.58E-n~

Standard 15 0.902 0.0466 3947 1.1Be·OS 0.897 0.0476 4055 1.1SE-OS 0.900 0.0472 4001 1.18E-1

Compaction a 0.975 0.0264 9629 2.74E-06 0.967 0_0278 8951 3.11E·06 0.971 0.0271 9290 2.92E·1 Soaked 7.5 0.959 0.0383 6555 5.54E-06 0.956 0.0383 6281 a.10E-06 0.958 0.0373 6418 5.82E-061

15 0.920 0.0472 3837 1.23E-05 0.899 0.0530 3900 1.36E-05 _L----L.-._. O.~~ -2~§.!- ~~=?~§.~05_

Speir Witczak

LIST OF FIGURES

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure II

Figure 12

Grain Size Distribution Curves

Dry Density Versus Rubber Content Results

Optimum Moisture Content Versus Rubber Content Results

CBR Values at Optimum Moisture Contents

Sand Coefficient of Permeability Versus Rubber Content

Kl Value Versus Rubber Content (Conventional Model)

K2 Value Versus Rubber Content (Conventional Model)

kl Value Versus Rubber Content (Universal Model)

k2 Value Versus Rubber Content (Universal Model)

k3 Value Versus Rubber Content (Universal Model)

Combined Observed Mr Versus Predicted Mr Analysis - Conventional

Model

Combined Observed Mr Versus Predicted Mr Analysis - Universal Model

I

~ I ! ! i I 1 I ! 1 !

~ I I I , I

I ; J I I

I : I I I ,

i I

~ I I

i--- I , - I , I , :

! I I , , I "- I --..... I I I "- I ....... I

I I I i ! I ........ I , , I I~ : i '\ I

, ! \ : 1\ I ,

, I !

\J ! ! \ , i : ,\ , ! I ;

I I

\\1 , , I I , I I i~ I I I I I I ! I , I

! I ;

"- \ "- \ ;

I I i I ; '\ I

I I -:-- j i I \ : I ;

I I I : _\1 I ! I I

I I I

, i .~ i I , i

i , I I , , I ! !

I , i ,

i

i\~ i I ! I I

, I I

I i I I i 1 I , I I I , : , I ,

I I ! I

I I ! I I I I I I i I ; i

I ! ! I I I I I I I I I I ! I I i I I

I I I 1 1 I I I !

I I I I I

I I I I

a ~ a ~ a a a a a a a a c:i a c:i a c:i a a a a '" co .... eo In .;- '" N ~

~

"'" BU!SSBd % '" ~ N .;;; B ~~

~ r"

,

i , , , !

,

i I ;

i !

, i I

I

' .

I

I

! ;

! ,

I I

~ a

a a

a ~

~

a

m In Ql

I I ;:. .. ::J U c: 0

~I ';::

"C ::J c: ..c

'" ';:

'" \

UJ ~

N In

iii

t is

'" Ql " '"

N

iii U5 c:

'(tj !Il .. q; t!l C!l

t ..... w a: :::l t!l u:::

Speir " Witczak

1 Ib/lt - 3 ~ 0.157 kN/m - 3 , 150

, ____ .. ~ __ r_· _____ ·_____ .-_. ---~.----'-' _.---.-.-- --- , [ , ,--,--

~ - ___ I __ _

140 -I-I--+---jf--""""!-.~---I--'~--I ---,-,--,--1---1--1--1-'--'--'

1-1---1_1 __ 1 __ --- ---- - 1- -I---I--I-I--I--I-f--------l---·--

130 ; ____ • ___ , __ '---I ------ I - ---. ~ - --- ---

iii • .::: __ • __ • ___ , __ 1-----1 ----1---, '-._- - -- -_.- --_.

Iii -"

.- ---- -- ----_. ---- - - ---~ 120 +-1 1-1- 1-1---.---.------'iii t:

" a >-~ a

__ -1---' ___ , ___ ' ___ ' __ ' ___ 1_. ___ ' -.- -, ---

110 +-/ /-/-1--1--.-,--'----'----- ____ . ,- ·--1 ---1--'---' --'---I~--'--'----

-

100 -I-I-

--.----,--.-- ,. __ .- ,-----.~ ____ -I

90 +-ll-_L_-'-_I-_I_-L __ ~ ___ · -

o 3 6 9 12

Rubber Content ('Yo) ----------------

• GAB Mod Compaction -=- GAB Stand Compaction

--+-- Sand Mod Compaction ----:- Sand Stand Compaction

FIGURE 2 Dry Density Versus Rubber Content Results

15

••

,

Speir Witczak

~ ... c

~ o

U CD

15 f-----I--f---l--l-.-.-.-.------.-.-..

"-- ---,~----.-----. --- .... ---.. [----I·- r - r-·-r -----.'-'- ---I-I·-l--··---

1-.

1.=1=1--1 1 I---

!--.. I--- f--. l-- ,... l-- f----

f--t---II--I-I---f---l-l--f-----I--+-+--+--I--f----I-I-I-I-I- I-I-·I---I--·-I·-I--I--I--I-+--+--f---l--l--l f----f----f----I-I--I--I-I--I--·I-·--1--1-I--I--I-I--II---t--

9 ·f..-I -l--l---- 1-

~ f----l--

1-I- 1- t--

- - _._---------------------_ .. ----- _.- _ .... _---- --. ------- .. -EEE~ffE~--- -.--.--. --- -" .--- ._-- --"8- - §§

3 - - - --.--' -- - .. ------ -.- ... --- --' --- ... --. - .-. - - ._' ... -------,-- ---------- .----- -_ .. _--- _._---.-

I~~ 1---1--1~-·-· .. -.. --··---·-· .. ·-----.f-__ ~ ___ ,_,_,. ___ ,_. ___ ,, ____ J .--.-

• __ • __ • __ • ___ • __ ..... ___ ._. __ 1 __ •• ·--·=~:·::-~~=~~--:~~~~I33 EEEEt=,

o _I !. I--L....-l·----·- .. -... -........ - .. - I---'--'-~--·--

o 3 6 9 12 15

Rubber Content (%)

• GAB Mod Compaction -=- GAB Stand --+-- Sand Mod Compaction -- Sand Stand Compaction Compaction

FIGURE 3 Optimum Moisture Content Versus Rubber Content Results

Speir

'. ,Witczak

"I

~ a: m U

100

90

80

70

-.-.-[---r--,--'----r---r---r"--y---

-- -, ---,---,---~--I---I___I---t--l.---

-'---f---' ---,---,---.---,---, ._--

60 -I-~-l-t--- - .. - .... -. '._-.- ,----

50 ?:::::::!:-...

_n ~

--f--40

30

20

10

0 1----'----'--'--' ' _____ L_ .. ---' _. __ 1 ___ 1 n. , ___ ... L __ , ... -+---- L , -.-1. __

0 3 6 9 12 15

Rubber Content (%1 ------.--------~------

GAB Mod Camp, ---0-- GAB Mod Camp, GAB Stand -GAB Stand As-molded Saturated Camp, As- Camp, Saturated

molded

Sand Mod Camp, - Sand Mod Camp, 0 Sand Stand -Sand Stand As-molded Saturated Camp, As- Camp, Saturated

molded

FIGURE 4 CBR Values at Optimum Moisture Contents

.><

'" ~ N .- (J " ~ 0. .-rn$

, !

i I

I , I ,

I I , ! I I

I

I

I

~

0 0

1 \

I 1 I

1 \

I I I I I I

I

I 1

1 I \ I , I ,

! , I !

! , I

i

! i

, !

i i I

I i , , I

i ! , ,

I I , ,

I

\

I I I

I I

I I I I

co 0 0 ci

I

i 1 1 i\ , , I ,

I I I \ I , I I

I

! I 1

, \ I I I

i I I I

I \ I I

I I I

\ I i , I

, I I i I I I I I I

! I I I I , I

\

I I I I !

I i I

i I

i i I

, ! I I

!

I i I ,

I I I

i I , ;

, ;

i I I , I ,

! ! i , ! ,

,

I i , I I

I , i I I i

;

I \ I I I

I ! I

I I 1 I I

I I I i I i I

, I 1 ! !

I I I I , I

I I

I I I I ,

<0 ... 0 0 0 0 0 0

(:>as/w:>1 'J{ ' ... ~!I!qBaWJad 10 ~ua!:>!llaoo

I I I ,

! I I

I I

I I ! , I

i

I N ... ~ c:

$ c:

1\ I 0 U

\\ I I

i \ i I ,

I

-QJ .c .c :::l

0::

i \ , i i !

I \1 I

rtl :::l rtl

'" -QJ

t< > \

, i

, I

i I

I I I

I ,

i 1 I

i

,

I

~ >-c: :!: " :E ~

c: 0 <tl U OJ ~ E " -.0 OJ .0 c.. " a: -<0

0 ... c: OJ

I

I I '0 -= -i ! QJ 0 U

I i I I \ I

'C c: <tl

IJ)

C'l U'I W 0::

I I I :J Q

\ I LI.

I I o

N 0 0 0 ci

"I

Speir Witczak

'u; .e-., " iii > ~

""

10000 '--~--l~-~-.--r--r---- .. -'--'.'-- - -.-... ---- ." _ .. ,------ llb/in - 2 = 6.B9 kN/m - 2 -1 I I .,------

--__ -'"---1 .• --1- __ 1 ___ '_ -- ,._--

BODO ~--I-I--I-~-'--'--

---I----l---'·,--·,----, ._-

6000

4000 -1--1- -- ---f~~--f"-..;~

" 2000 1 1---1 I~-

---,---1---'---'----'--'--- ,------... --1----. -,- --- ---I----I--I-I-I-f----·-·-

o -~~--'--I __ - L. __ I __ _

o 3 6 9

Rubber Content (%) .. _--- -- ------- - .-

• GAB Mod Comp, --0-- GAB Mod Camp, As-molded Saturated

• Sand Mod Comp, As-molded

" Sand Mod Camp, Saturated

--- GAB Stand Camp, As­molded

---- Sand Stand Camp, As­molded

12

-:- GAB Stand Comp, Saturated

-- Sand Stand Comp, Saturated

FIGURE 6 K1 Value Versus Rubber Content (Conventional Modell

15

,

Speir Witczak

0.7 -I -r---' .. _- ",-'---r-"--I- .. ~-••.• -----.( -- j'-- -'r~ "--T- ---,-·----f"

~_I 1--1-1--1- - 1- ---1--1 ---1--

0.6 -l-I ~-I--- ,--- ,- 1-- I--I-I-I-I-I--~--I-I--

17 0.5

./ " ., " iii > <"II :.: 0.4

- ___ 1 __ 1--1- --, ----- ,----

~-I---I-I-I--I-I-I-----I----I---- --1-

0.3 i I-I-I--I----f--I---I-- --1--1----1----,-- 1-1--1-1-1-1-1-1----

-1-- ---., ---

0.2 _____ l ... _ __ -1_.l __ L_I __ L_L-__ L--_' --

o 3 6 9 12 15

Rubber Content ("!o) --- ------------

-- GAB Mod Compo -=- GAB Mod Camp. GAB Stand -:- GAB Stand

As-molded Saturated Camp. As- Camp. Saturated molded

Sand Mod Camp. --0-- Sand Mod Camp. 0 Sand Stand -=- Sand Stand

As-molded Saturated Compo As- Camp. Saturated

molded

----

FIGURE 7 K2 Value Versus Rubber Content (Conventional Modell

~

Speir Witczak

10000

8000

1 Iblin - 2 = 6.89 kN/m - 2 -,---,----,--,----, i --,----

f--f--t--t--f--'_-'-"'-- ,.- .. _- ,---_ .. -- ,-- .. ------1--- '''' ----,---- --- '--i-f 1-1--1--

-l-I 1-1-1 1--1--1-1-1---1--1--1--1---+ 1--1-1---1---1-

f--f--I--I--I._I--I--I---I-- -1-1-1 .. -- 1----1--.. -/--1-- -1-1--1--1-1---1---1--1--,--

'iii 6000 -1-+-·1--1--'--'--'--'--" --.- ,----,-­.E-m

" iii > .... .>: 4000

a

Jr-"""k,--I---j---'---'--'--'--' ,--,---.,-_., - ---" .• ----- --,---. '---1-1-1---1-1-1--1--1-

.-~- ----.-.'-_'.-.'

-__ , ____ ,.,--,---,---,---,---,---,---1---,---

-J-_L---1._-'-.--!L-- _---'_1-__ L_1.._I_.1 _ .. 1 ___ L ..... ~ ---'I---L L __ I_-L---L--L-l __ -1

a 3 6 9 12 15

Rubber Content (%)

GAB Mod Camp, --0-- GAB Mod Camp, GAB Stand -GAB Stand

As-molded Saturated Camp, As- Camp, Saturated molded

Sand Mod Camp, --0-- Sand Mod Camp, 0 Sand Stand --0--- Sand Stand

As-molded Saturated Camp, As- Camp, Saturated molded

FIGURE 8 k1 Value Versus Rubber Content (Universal Modell

Speir Witczak ,

,

N

""

1.200

1.000

0.400

0.200

0,000

,--,--,---,----,--r r-~-I--·-·--r·------I---r· - -- --·-----I-----r--- I-----I---r--r r--~--r--'~

~"-'--' .. 1··----1 ..... , ... _, ... _-\_ .. -·-1-1--1-1--1--1----1-1--

--l-- 1-_·-

. --. 1·-·--·"-· ·-1-"

1--1-1-·1---1---1-1-1-- ,-.. - --I-·-I--I-j··--I--I---

~-I--I__l-I--l-I-I--I---+·--I- _.--·-.1--1-1-··+·-1-1-1-1-1--1-1-1--1

. --/--1--1-- 1--·-1--1---'·· - , . -..

-I-l-f-I--I--I--I---I-I-·-I- I-- I--.I -·-/- ·/--/---/-1-1-1--1--1 I-I

1--11--1--1-1-' .-.. - -1- -1--1-·1--1--1-1---1-· .-

.1 l_l_1_·.I __ .J_· __ I __ I . .1. _1. __ L __ ., __

o 3 6 9 12 15

Rubber Content (%) ~----.--- .. ----...... - ... _.-....... __ ._-_._-- -----_ ... -

GAB Mod Compo --oU--- GAB Mod Compo GAB Stand -:- GAB Stand

As·molded Saturated Compo As· Compo Saturated molded

Sand Mod Compo --Sand Mod Compo I Sand Stand ~Sand Stand

As-molded Saturated Compo As- Compo Saturated molded

FIGURE 9 k2 Value Versus Rubber Content (Universal Modell

,

Speir Witczak

'" " iii > M .><

, . ,

0.000 ,--,._---- ··'----T------ ,-- -----', --'---r--~ ----- ..

__ ' ___ 1 __ 1-------1.---1---, ____ I.

-0.100 _____ -1-- _1 ___ .1 __ 1 ___ ,--. ---.---,---. ---I.---'---~I--,-

________ '. ___ 1_. ___ 1 ___ •• ___ •·· ____ 1 __ ···-...j-I---I- 1--1--'-'--'-'

-0.200

I-~--

-0.300 1- ~

.. ---I.-:t:: I I -,. ';J.:: , , '!

I--I-I-I-+--I---I--I-I--I-~-I----I ---I ----I.--I-t-t-t-r-r--I-I-==r

-0.500 I 1--1_1--J __ .L_I ___ I_.I _____ 1 ____ '-__ .I -1--·- L.--l_I_L-I---' __ .L_L-L __

o 3 6 9 12 15

Rubber Content (%)

GAB Mod Camp, --C_ GAB Mod Comp, GAB Stand --:- GAB Stand

As-molded Saturated Comp, As- Camp, Saturated molded

Sand Mod Comp, -- Sand Mod Comp, a Sand Stand --<>- Sand Stand

As-molded Saturated Comp, As- Comp, Saturated molded

FIGURE 10 k3 Value Versus Rubber Content {Universal Modell

r.

0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

" '" '" .,. " N

IIsd) JW pal:llpald

...

,

Speir Wilczak

---.---~--

70000 -r- ---,------ r . - -I

--

60000 1 1 ----il---

1---- -1------\----1----1--\----

50000 -I \---1 f-----I--I-- 1---+-

1lb/in-2 6.89 kN/rn-2

----~ •

; 40000 R2 = 0_990

I/i'l ....... t--- -- = 1350 . . , • •

-- ----- - ; ,~ = 0 _ 099 -- /SY_1

I - - --- -,------,-.- .,._-- f-----

:iii ~ • ~ • 30000

• .t

-- j- - -

20000 1---- ----1---1---1---\----1

10000

... _-_-1. ___ ~, __ .

o L'J _____ 1 _____ L _____ I ____ L ___ \ __ L-_I

o 10000 20000 30000 40000 50000 60000 70000

Actual Mr IP5j)

L --------- --- -----_._-- --'"-----

FIGURE 12 Combined Observed Mr Versus Predicted Mr Analysis - Universal Model

l

~-