Roller-Compacted Concrete
Pavements Home >
Roller-Compacted Concrete >Research in Progress
Refinement of
Roller-Compacted Concrete Pavement Fatigue Design Life Curves
Principal Investigator: Paul Okamoto/Matt Sheehan
Objectives
Roller Compacted Concrete (RCC) is a zero slump portland cement
concrete mixture that is placed with an asphalt concrete paver and
compacted with vibratory and rubber-tired rollers. The concept of
RCC pavements originated in Canada during the mid-1970’s for
use in log-sorting applications where roads are subjected to relatively
heavy loads. Construction of RCC has been expanded to roadways and
to container ports, rail terminals, truck terminals, and industrial
yards where the pavements are subjected to large forklift trucks,
straddle carriers, log stackers, and mobile cranes.
Thickness requirements for RCC pavements are established to minimize
flexural fatigue cracking over the design life of the pavement.
The number of load repetitions to cracking is determined from a
fatigue relationship that incorporates RCC design strength and stress
due to vehicle loadings. Commonly, the PCA developed RCC relationship
fatigue is utilized in RCC pavement design. This fatigue function,
developed by PCA in the mid 1980’s, was based on fatigue testing
of 23 beam samples sawed from RCC test slabs. The slabs were constructed
using RCC mixes commonly utilized in dam construction. The fatigue
function was based on RCC mixes with high coarse aggregate contents,
low cementitious contents (approximately 3.2 bag at 17 to 20 percent
fly ash substitution), and water cementitious ratios of 0.46 to
0.53. These mixes, representative of those used in dam construction
in the early 1980’s, are not representative of mixes used
in present RCC pavement construction. The RCC mix formulations have
changed and the methods and/or equipment used in RCC batching, placement,
and consolidation have changed since the PCA RCC fatigue relationship
was developed.
The objective of this project is to refine fatigue curves used
in establishing RCC thickness requirements and to develop strength
testing procedures for quality assurance. Specifically, the project
objectives are to:
1. Review fatigue-cracking performance data of RCC pavements that
have been subjected to large numbers of load repetitions. It is
important to note that it will be difficult to obtain the following
information: RCC strength (specified or tested), nominal RCC thickness,
magnitude and type of wheel loads, number of daily load repetitions,
types of distress (cracking, spalling, shattering) and how this
distress is progressing).
2. Fabricate RCC beam specimens utilizing different sources of
coarse and fine aggregates, different cement contents, and different
water-cement ratios. Mixes will be representative of those used
in present day RCC pavement construction. Beams will be tested to
establish the flexural strength and endurance limit (fatigue life).
Mix designs tested will be consistent with what is done in Canada
and in the US. In Canada, some beams and cylinders should be based
on SEM’s mix design procedure and some others on the method
used in Western Canada. Data will be used to refine the PCA RCC
fatigue curve utilized in establishing thickness requirements. (Cylinders
will also be fabricated and tested to determine the RCC splitting
tensile strength, compressive strength, and modulus of elasticity.
Cylinder strengths will be correlated with beam strengths. Correlations
will then be used to establish construction quality assurance procedures
utilizing cylinder strengths. Data will also be used to establish
strength and stiffness characteristics and other structural indices
that can be used in current and future pavement design procedures.
3. Modifications as a result of the above research will be incorporated
into the RCCPave Program (by means of downloadable patch to the
program with new fatigue curves).
Significance of the Project
Thickness requirements for RCC pavements are established to minimize
flexural fatigue cracking over the design life of the pavement.
Fatigue performance is primarily a function of RCC flexural strength,
slab support (modulus of subbase/subgrade reaction, k), load-induced
flexural stresses, and number of load repetitions until cracking.
For a given slab thickness, the designer will first calculate load-induced
stresses as a function of thickness and design modulus of subbase
support (k-value). The number of load repetitions to cracking is
then determined from a fatigue relationship as a function of the
RCC design strength and calculated stress. If the allowable number
of load repetitions calculated from the fatigue relationship is
greater than the design number of load repetitions, the slab thickness
is sufficient to minimize cracking over the pavement design life.
The process can be repeated, increasing or decreasing the slab thickness,
until the minimum required thickness is achieved.
For a specific project, the designer specifies or estimates the
RCC flexural strength and k-value. Load-induced flexural stresses
as a function of thickness and design modulus of subbase support
can be calculated using PCA’s RCCPave2000 program. The remaining
design parameter is the fatigue function relating the number of
repetitions until cracking as a function of the calculated stress
and specified strength. Commonly, the PCA RCC fatigue is utilized
in design. The fatigue curve, developed from RCC mixes typically
utilized in dam construction, may not be applicable for RCC utilized
in present day pavement construction.
Utilizing a fatigue curve that is not representative of RCC fatigue
performance can significantly affect thickness requirements. Based
on field observations in Canada and in the U.S., it appears that
the fatigue relationship used in the current PCA developed RCC pavement
thickness design procedure may be too conservative and does not
recognize the improvement in the quality of RCC pavement construction
that have taken place in the last 10+ years. If calculated thickness
requirements using an inappropriate fatigue curve are too conservative,
the pavement thickness may be over designed and material costs are
driven up to a point that makes RCC less competitive.
PCA’s RCCPave2000 was used to develop the following example
when the current fatigue curve is conservative by 10 percent. A
10 percent shift assumes that for a given number of load repetitions
that the allowable stress is actually 10 percent higher than currently
allowed. Using the default values in RCCPave2000, for a load-induced
stress of 335 psi, the required thickness is 15 in. When the allowable
stress is increased by 10 percent to 370 psi, the required thickness
is reduced by 14 in. For this example, if the allowable stresses
in the present PCA fatigue curve are conservative by 10 percent,
the required thickness is conservative by 6 percent. On large industrial
projects, a reduction from 15 to 14 in. corresponds to a substantial
reduction in construction and material costs.
The refinement of the PCA fatigue curve can result in more optimized
design thickness for RCC pavements. This project will refine the
current fatigue curve using RCC mixes more representative of those
used in current RCC pavement construction in the U.S. and Canada.
Project Description
The project will consist of the following four tasks:
- Task 1 – Review of RCC Fatigue Performance
- Task 2 – Fatigue Testing and Refinement of PCA Fatigue Curve
- Task 3 – Development of Inter-Strength Relationships and Quality
Assurance Recommendations
- Task 4 – Final Report
Task 1 – Review of RCC Fatigue Performance
This task involves a review of heavy industrial RCC pavement fatigue
performance. The PCA RCC fatigue curve was developed from laboratory
testing of RCC beam specimens. The curve was never field calibrated
with fatigue cracking performance. A literature and performance
documentation review will be made to establish whether the current
fatigue curve is conservative or not. This effort will involve gathering
of information from recent surveys and discussions with engineers
at heavy industrial facilities. Information will include, if available,
RCC strength (specified or tested), nominal RCC thickness, magnitude
and type of wheel loads, number of daily load repetitions, types
of distress (cracking/spalling/shattering), and how fatigue cracking
distress, if any, is progressing. Performance data will also be
used to evaluate the refined fatigue curve.
We anticipate that much of the data for Task 1 would be forwarded
to CTL by PCA/CAC field engineers at no cost to the project. Other
data will be gathered from phone conversations with engineers at
industrial facilities as well as consulting/design engineers involved
with RCC pavement design. Also, a selected number of field visits
will be made to obtain more in-depth information, to document the
fatigue performance of these projects, and retrieve core samples
for strength verification. It is anticipated that site visits will
be limited to those where a RCC pavement has exhibited fatigue cracking
failure.
A technical summary documenting the results of Task 1 will be prepared
for review. This technical summary can be used to identify whether
any fatigue cracking problems exists (present fatigue curve not
conservative) or no fatigue problems exist (present curve conservative).
Task 2 – Fatigue Testing and Refinement of PCA Fatigue
Curve
Task 2 will involve fabrication of beam specimens to be tested
in fatigue. RCC mixes with different coarse aggregates, cement contents,
and water-cement ratios, representative of those used in present-day
construction of RCC pavements, will be batched at CTL. It is anticipated
that the fatigue curve and inter-strength correlations will be based
on approximately four representative RCC mixes. Using commonly accepted
consolidation procedures, RCC will be consolidated into steel beam
molds with electrically driven rammers. Beams will be moist cured
and tested under repeated loads subjected to different stress magnitudes.
It is anticipated that approximately 40 beams will need to be tested
in fatigue to develop a statistically significant fatigue curve
that can be utilized for design. The number of load repetitions
will vary from 1000 to 500,000 load repetitions.
Fatigue data will be analyzed to develop a refined fatigue curve
at various levels of reliability. Data from the original PCA fatigue
curve may also be incorporated. Developing a set of fatigue curves
(Note: There may be the need to develop two separate fatigue curves
(one for Western Canada/US the other one for Quebec region) at different
levels of reliability will give designers an option to incorporate
different levels of risk depending on whether pavement is located
in a critical location or not.
Task 3 – Development of Inter-Strength Relationships
and Quality Assurance Recommendations
Companion cylinders, made from the RCC fatigue mixes, will be tested
to establish correlations between beam flexural strength and cylinder
compressive/splitting tensile strength. A correlation will enable
cylinder, rather than beam tests, to be conducted during construction
quality assurance strength testing.
The modulus of elasticity will also be determined to establish
a correlation between RCC compressive strength and modulus of elasticity.
Data will be used to establish strength and stiffness characteristics
and other structural indices that can be used in current and future
RCC pavement design procedures.
It is anticipated that approximately 100 cylinders will be fabricated
and tested to establish compressive strength, splitting tensile
strength, or static modulus of elasticity.
Task 4 – Final Report
A draft final report will be prepared documenting all activities
conducted in this project, including the results of the fatigue
study, recommendations for refined fatigue curves (different levels
of reliability), recommendations using of cylinder strengths for
quality assurance, quality assurance strength testing guidelines,
and RCC strength/stiffness parameters for use in future RCC pavement
design. The report will be finalized based on review comments received
from PCA.
Schedule
The project duration will be 24 months. The work schedule is summarized
below:
Task 1 – Revised to February 28, 2004
Task 2 – Revise start date to January 2004
Task 3 – Revise start date to January 2004
Task 4 – December 31, 2004
|