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