Dennis Polhill

Deflection Analysis and Case Illustrations of Thin Asphalt Pavements for Overlay Design

April 9, 1984

Frank Meyer, PhD, P.Eng.
PMS Group, Cambridge, Ontario

Ralph Haas, PhD, P.Eng.,
Professor & Chairman, Dept. of Civil Engineering
University of Waterloo

 

Dennis Polhill, MSCE, MPW, P.E.
PMS Group, Denver, Colorado

 

Paper Prepared for
Presentation to
The Annual Meeting of
The Association of Asphalt Paving Technologists
Scottsdale, Arizona
April 9-11, 1984

ABSTRACT

Lack of structural adequacy is often a major reason for the rapid deterioration of a thin asphalt pavement. Deflection testing and analysis provides a sound basis for assessing the degree of such structural inadequacy and for designing overlays of sufficient thickness to strengthen the pavement.

This paper describes the basic role of overlay design as a project level activity within pavement management, defines the factors involved in deflection testing, describes how deflection data can be analyzed and used in overlay thickness design, and provides a specific case illustration.

Deflection testing involves three basic considerations: (1) type of equipment, (2) the testing program itself and (3) analysis and use of the data for overlay design. The paper identifies the most common pieces of equipment used in North America, and some of their features. As well, a suggested test program for different classes of highway or streets is given.

The analysis of deflection data for use in overlay design depends on the design method to be used. This can range from an empirical approach where overlay thicknesses are to be sufficient to reduce the surface deflection to a maximum tolerable value, which itself is a function of the traffic level, to the theoretical or mechanistic where material properties and elastic layer models are used. While the latter approach is desirable, it has some problems with thin asphalt pavements; consequently, the empirical approach is currently more applicable in this case.

An example is provided in the paper which uses empirical procedures but which is equally applicable to a mechanistic procedure in terms of results. The example illustrates how a designer can use computer graphics and analysis in an efficient way to capture the deflection variation along a section of road, and between lanes, in an efficient way so that a final, practical and balanced overlay design can be determined.

INTRODUCTION

The United States and Canada have a large mileage of thin asphalt pavements. These pavements can deteriorate fairly rapidly and require rehabilitation after relatively short service lives.

Rehabilitation, including overlaying, may be required for one or more of the following reasons:

1. Inadequate structural capacity for current or expected future traffic loading

2. Unacceptable level of service in terms of Present Service ability Index (PSI) or Riding Comfort Index (RCI)

3. Unacceptable level of surface distress

4. Unacceptable level of safety

5. Unacceptable costs to the road user

6. Unacceptable maintenance costs.

In the case of thin asphalt pavements, lack of structural adequacy is usually a major reason, which itself is responsible for accelerated loss of serviceability, surface distress, etc.

Several methods exist for assessing structural adequacy, both destructive (i.e. coring) and non destructive in nature. The latter usually is carried out by deflection testing, with several devices commonly used in North America. Consequently, deflection testing and analysis represents a sound basis for overlay design, especially in the case of thin asphalt pavements.

But the way in which deflection data can be and is used for overlay design varies widely, ranging from the relatively simple and empirical to the theoretical where back calculation of material properties is used with layered elastic models. The latter, often termed a “mechanistic approach” is attractive in that it allows new materials, varying load limits and configurations, varying materials properties, etc. to be explicitly recognized. However, it can also have limitations, as subsequently discussed.

Scope and Objectives of the Paper

This paper is concerned with the use of deflection testing and analysis to design overlays for thin asphalt pavements. More specifically, the objectives are as follows:

1. To identify the role of overlay design in pavement management

2. To define and describe the factors involved in deflection testing

3. To describe how deflection data can be analyzed and used in overlay thickness design, and

4. To present specific examples of the foregoing.

OVERLAY DESIGN AND PAVEMENT MANAGEMENT

Much has been written on pavement management, and there are two books available (1,2) plus many published papers. It is generally accepted that pavement management is actually carried out at two distinct but interrelated levels. One is the network level, where the end result is priority programs of rehabilitation, maintenance and new pavement construction for the various years of the program period. The other is the project level, where the end result includes detailed designs for the projects that come “on stream” from the network program, plus construction and periodic maintenance. Figure 1 illustrates these two basic levels of pavement management.

Overlay design of course is a project level activity. It requires more detailed testing, including deflection, than the network level. For example, a network level field inventory may involve very limited deflection testing, if at all [There is a lack of agreement in the pavement engineering field as to whether deflection surveys are needed at the network level. For example, Arizona and New York do not use them; Utah and Idaho use them. While good arguments can be made for either situation, the authors of this paper support the use of deflection surveys at the network level because they feel the “quality” of the information gained far outweighs other, indirect measures of structural adequacy (such as layer thicknesses), and because of its value in identifying needs and developing rehabilitation strategies.], while deflection testing for a project level overlay design may be carried out at 50 m or closer intervals. Karan, et al (3) have shown that an approximate overlay design of uniform thickness for an entire section length is sufficient at the network level, and that subsequent detailed testing and design at the project level, when it has come on stream (i.e., it’s designated year in the program has arrived), can “fine tune” the design to result in padding and level up where needed along the section length, varying thicknesses, detailed quantities, etc. that are required for the actual contract.

The following sections in this paper assume that the first network level of pavement management has been carried out, that a project has been programmed, and that the next step is detailed testing, analysis and design.

DEFLECTION TESTING

Of the non-destructive methods available for measuring pavement response to load and hence evaluating structural adequacy, deflection testing is by far the most popular. Deflection as a structural measure has many advantages. It is relatively simple, generally proportional to load and a good indicator of how the pavement will perform. However, deflection measurements also have some limitations, including wide variations with season and temperature which must somehow be represented for design purposes, the fact that maximum deflection under load may not adequately capture the structural characteristics of the pavement and that deflection is sometimes not well related to fatigue and permanent deformation damage.

If deflection testing is to be used, the following questions must be answered:

l. What type of equipment is most suitable?

2. How should a testing program be set up (i.e., factors to consider, one lane or both lanes, spacing of measurements, etc.)?

3. How can the data be analyzed and used for overlay design?

The first two questions are addressed in the following paragraphs, while the third question is addressed in the next section.

Regarding equipment, the most commonly used types in North America are the Benkelman Beam, Dynaflect, Road Rater and Falling Weight Deflectometer. The various features and attributes of these pieces of equipment were extensively considered in two FHWA work-shops on pavement management in Phoenix and Charlotte in 1980. Table 1 provides a summary listing (4). It is not the purpose of this paper to try and define which is best on an absolute basis, because each has advantages and disadvantages which vary in importance with the individual user. The Dynaflect is used for illustrative purposes in this paper, primarily because of its widespread use and experience with it. However, the principles employed and type of results subsequently shown should be similar for any of these four pieces of equipment.

The operation and theories associated with the devices listed in Table 1 are described in detail in many publications, including Ref. (5,6).

TABLE 1 – STRUCTURAL EVALUATION EQUIPMENT

After Ref. (4)

	INVENTORY
	OR DESIGN	Cap.	COST OF	MAN-		RELI-	REPRODUC-	PRODUCT-	AUTO-
EQUIPMENT TYPE	(I OR D)	Cost	OPERATION	POWER	SAFETY	ABILITY	IVITY	IVITY	MATED
Benkelman Beam	I/D	Low.	High	2 – 6		Good	Fair	Medium	No
Dynaflect	I/D	Low.	Med.	1 – 6		Good	Good	Med/Hi	Avail.
Road Rater	I/D	Med.	Med.	1 – 6		Good	Good	Med/Hi	Avail.
Falling Weight Deflectometer	D	Med.	Med.	2 – 6		Good	Good	Med/Hi	Avail.
Cox Meter	D	?		1 – 6		Fair	Good	Hi+	Avail.
California Deflectometer	D	High		2 – 6		Satisf.		Medium	?
LaCroix Deflectograph		High		2 – 6		Satisf.		Medium	Yes
WES Vibrator		High		3 – 6		Fair	Good	Medium	No
FHWA “Thumper”	Research	High +		3 – 6		Fair	Good	High	?
Plate Bearing		High		3 – 6		Satisf.		Low	No
Field CBR
Laboratory Tests

Test Program and Procedures

The development of test programs and procedures for deflection testing is usually influenced by the following factors: (1) testing budget available, (2) type or class of highway or street – i.e., free way, arterial, collector or local, (3) variation of subgrade and drainage conditions along the section length (4) number of lanes, (5) volume of traffic, (6) length of the section itself.

For thin asphalt pavements these factors can assume particular importance and it may be necessary to do some “tailoring” of the testing program to the particular conditions at hand. However, for thin asphalt pavements and average conditions, the program shown in Figure 2 may be used as a guideline. It is based on a background of experience with many thousands of miles of both rural and urban pavements.

The costs will of course vary somewhat with the interval spacings shown in Figure 2. Larger intervals could cut the costs somewhat, but the decrease would be marginal and the quality of information loss could be substantial.

DEFLECTION ANALYSIS AND THICKNESS DESIGN

Deflection test results can either be used almost directly, with a minimum of analysis, in designing overlay thicknesses, or they can be used to “back calculate” material properties using mechanistic analyses. A considerable number of such mechanistic analysis methods exist, some of which are listed in Table 2, along with references. The use of mechanistic procedures for overlay design has been described in many publications and is incorporated, with a considerable number of variations, in the overlay design procedures of various agencies. Epps and Hicks have provided a good summary of typical approaches and methods (31). While the advantages of such mechanistic procedures are many, as previously pointed out, and while improvements in technology will undoubtedly lead to their use on a more widespread and effective basis in the future, they currently have some major limitations with thin pavements. The results (i.e., material properties, stresses and strains, deflections) are often quite unrealistic and the reasons probably include some of the boundary condition assumptions used in the layer theories.

The empirical deflection based procedures thus find more widespread application in the case of thin asphalt pavements and are used in the following paragraphs and sections. Again, however, the types of results shown could be the same for empirically or mechanistically based design procedures.

The empirical, deflection based approaches are typified by The Asphalt Institute’s procedure in their 1969 MS-17 Manual (32), The Roads and Transportation of Canada’s (RTAC) procedure (1), and the British TRRL method (33). They all are based on providing sufficient overlay thickness to reduce the surface deflection to some standard or maximum tolerable value, which is a function of the number of equivalent single axle loads to be carried.

TABLE 2 – TYPICAL METHODS – FROM DEFLECTION TO MATERIAL PROPERTIES

METHOD	COMMENTS	REFERENCES
SWIFT	2 layer system, elastic	7
VASWANI	equivalent 2 layer system	8,9,10
ODEMARK	equivalent 2 layer system	11,12
WISEMAN	2 layer system	13,14
ULLIDTZ	3 layers, equivalent system	15,16
IRWIN – (MODCOMP, NELAPAV)	8 layers, non-linear elastic model (stress-dependent)	17,18,19,
THOMPSON – ILLI-PAVE	stress-dependent finite-element	20
BUSH – CHEVDEF	4 layers, layered elastic model	21,22,23
DYNATEST – ISSEM4        – ELMOD	4 layers –elastic moduli3 layers – limited stress-dependent equivalent system	24
SHELL – BISAR	99 layers, layered elastic	25,26
PENN. STATE	elastic layer, 4 layers	27,28
MAJIDZADEH-MOD4	4 layers, elastic layer	29,30
KENTUCKY

Maximum Tolerable Deflection

The maximum tolerable deflection (MTD) is a function of traffic and various relationships have been developed by a number of agencies. Figure 3 provides some comparisons between various agencies in North America, England and Australia.

Correlation to Benkelman Beam

Since the empirical, deflection based procedures were developed with Benkelman Beam deflections, correlations must be established if, for example, Dynaflect is used. Figure 3 provides example correlations for several agencies. These can vary widely, depending on sub-grade soil type, pavement type, and climatic conditions.

It is desirable that an agency establish a correlation for its own particular area.

Also, there is usually considerable-scatter in the data whenever such correlations are attempted. This is quite understandable because the Dynaflect and Benkelman Beam are quite different in configuration, and what and how they measure (i.e., dynamic vs static). But the scatter often results in engineers believing one or the other device is deficient (which is not the case, for the aforementioned reasons) or that the correlation is useless. The latter is also not valid because the errors involved in estimating Benkelman Beam deflection from Dynaflect deflection are not inconsistent with other errors and variations in the overlay design and construction process (i.e., variation in deflection along the section length, errors in predicting traffic, and variations in constructed thickness).. Moreover, these errors should be random and thus have the effect of negating each other.

Overlay Thickness

When the Benkelman Beam deflection has been directly measured to be overlayed, (usually referred to as the design estimated from Dynaflect measurements (using the type in Fig. 4), then the overlay thickness required to reduce the surface deflection to the MTD value (which has been determined from a Fig. 3 type relationship) can be determined from the type of design curve shown in Figure 5. The overlay thickness is in inches of equivalent gravel and has to be converted to asphalt concrete. For example, if the Benkelman Beam deflection on the section were 125×10-3 in., and if the MTD were 0.070 in., overlay thickness required (Fig. 5) would be about 5.6 in. Further, if the equivalency were 2:1 for asphalt would be 5.6 : 2 = 2.8 in. on the section deflection) or of correlation existing then the of equivalent gravel. concrete to gravel, then the actual thickness = 2 3/4 in. of asphalt concrete.


Adjustment to Design Deflection Values

Two major adjustments are usually required to design deflection values:

1. Conversion of the (measured) design deflection to a peak spring value (because the design charts are based on this)

2. Temperature corrections.

Extensive studies in Canada (1) and elsewhere have shown that the ratio of fall to spring deflection values usually range between about 1.6 to 1.8. Thus, if a set of deflection measurements are made in the summer or fall, they should be multiplied by about 1.7 to convert them to peak spring values. However, this should be used with considerable caution because it certainly varies with region or climatic zone. For example, in Texas or Arizona, these ratios might be closer to 1.0.

Temperature correction graphs have been published by the Asphalt Institute (32) and RTAC (1).

CASE ILLUSTRATION

An actual example of a partially automated overlay design for a thin asphalt pavement can illustrate the approach and procedures out-lined in the foregoing sections.

It involves 2.825 km (1.75 miles) of two-lane rural road with an AADT of 5,000, commercial traffic volume of 5% and estimated growth rate of 7.5% per year. The average daily number of equivalent single axle loads in the design lane (DTN) for the design period of 10 years has been calculated as 226, with an accumulated total of 680,000 for the 10 years. Summary data for the project is shown in Table 3.

TABLE #3

SUMMARY INFORMATION FOR THE OVERLAY DESIGN PROJECT
NUMBER OF TESTS	57	AADT (BOTH DIRECITONS)	5000
MAN OF S-1	0.90	GROWTH FACTOR	1.53
STANDARD DEVIATION OF S-1	0.38	AADT IN DESIGN YEAR	3826
DESIGN DEFLECTION	2.41	COMMERCIAL TRAFFIC CONTENT	191.3
ANALYSIS PERIOD (YEARS)	10	ESAL EQUIVALENCY FACTOR	1.00
SEASONAL ADJUSTMENT	1.80	DTN	226.4
NUMBER OF LANES	2	TOTAL-DESIGN PERIOD ESAL	679082
COMMERCIAL TRAFFIC (%)	5	MTD (MILS)	1.45
GROWTH RATE (%)	7.5

For the traffic levels shown, a maximum tolerable deflection value of 1.45 miles is determined, using the Ontario criteria in Figure 3. This means that the actual deflection or design deflection must be reduced to this value by sufficient overlay thickness.

The field work consisted of the following: (1) deflection testing every 50 m (150 ft), (2) cores to determine existing layer types and thicknesses and (3) a surface condition survey.

Figure 6 provides a plan view of the roadway plus a variety of other information. The coded boxes on this plan view give estimates of the subgrade’s ability to support the loads. Also on this plan view are the core/bore locations. Above and below the plan view are the calculated overlay thickness recommendations and deflection profiles for each lane; the solid horizontal line on each of these plots is the MTD (adjusted by the seasonal adjustment factor) and the measured deflection profiles then appear along the length of the plot for each station. Whenever the deflection profile is above the adjusted MTD line, an overlay is required to provide some strength and reduce the deflection to the acceptable value. The number written above the rectangular box at each station represents the calculated overlay thickness in millimetres of asphalt necessary to reduce the deflection. For example, at station 0+300 in the eastbound lane 160 mm (6 in.) of AC is required to reduce the deflection from 2.2 mil to the acceptable MTD. The top and bottom sketches provide the Dynaflect deflection bowls which can be used to identify the source of the problem if overlays are needed. Obviously the recommended overlay thicknesses for each lane will have to be translated to practical values for the entire roadway width and for actual construction. This will subsequently be illustrated.

An examination of the core/bore data for the project indicated about 50-55 mm (2-2 1/2 inches) of asphalt concrete for varying thickness of granular 225-600 mm (9 to 26 inches) for the existing pavement. It also indicated a SW-SM subgrade, using the unified soil classification.

The surface distress report, Table 4, shows raveling throughout and various amounts of cracking. The ride was considered good throughout.

TABLE #4

RESULTS OF SURFACE CONDITION SURVEY

CLIENT: XXXXXXXXXXXXXXXXXXX
PROJECT: DETAILED SURFACE DISTRESS REPORT
DATE: JUNE 1983
SECTION: 0001
LANE: EAST BOUND LANE #1
LOCATION: TAYLOR BOULEVARD FROM MONA DRIVE TO GARDINERS ROAD

STATION	STAT SDI	PAVEMENT TYPE	RIP/ SHOV	RAV/ STRK	FLUS H/BL	DIST ORTN	EX C ROWN	EDGE CRK	ALLI GATR	POT HOLE	MAP CRAK	LONG CRK	TRAN CRK	RUT TING	STAT SDI
00+000	9.2	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	7.6	10.0	9.2
00+050	9.4	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	9.4
00+100	9.4	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	9.4
00+150	9.4	OLD AC	10,0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	9.4
00+200	9.4	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	9.4
00+250	9.4	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	9.4
00+300	8.8	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	4.0	10.0	10.0	8.8
00+350	8.8	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	4.0	10.0	10.0	8.8
00+400	9.2	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	7.6	10.0	9.2
00+450	9.4	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	9.4
00+500	8.6	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	2.6	10.0	8.6
00+550	9.4	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	9.4
00+600	9.4	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	9.4
00+650	9.0	OLD AC	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	5.6	10.0	9.0
00+700	7.8	OLD AC	10.0	10.0	10.0	10.0	10.0	4.0	5.1	10.0	10.0	10.0	10.0	10.0	7.8
00+750	8.8	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	4.0	10.0	8.8
00+800	9.4	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	9.4
00+850	8.8	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	4.0	10.0	8.8
00+900	9.4	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	9.4
00+950	9.0	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	7.6	7.6	10.0	9.0
01+000	9.2	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	7.6	10.0	9.2
01+050	9.2	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	7.6	10.0	9.2
01+100	9.2	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	7.6	10.0	9.2
01+150	9.2	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	7.6	10.0	9.2
01+200	9.0	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	7.6	7.6	10.0	9.0
01+250	8.2	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	5.1	10.0	10.0	10.0	7.6	10.0	8.2
01+300	9.2	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	7.6	10.0	9.2
01+350	8.0	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	3.4	10.0	10.0	10.0	10.0	10.0	8.0
01+400	9.4	OLD AC	10.0	3.3	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	10.0	9.4

Consideration of a11 this information suggests the road is in reasonable good shape, but without some strengthening in the not too distant future, it will deteriorate quite rapidly under those high traffic volumes.

The deflection profiles for each lane are basically duplicates in that they are low at locations where the cores identified bedrock (#’s 2 and 4) and high on the silty sand subgrade (#’s 1 and 3).

The profiles for the 2 lanes also peak at the same locations and drop off together, further substantiating the effect of the bedrock.

The rehabilitation recommendation that comes from the assessment of this data is a combination of an overlay and padding where appropriate. The road has not deteriorated (distress wise) to warrant removal of any large amounts of cracked areas before padding; normally this should be considered in all cases of badly distressed pavements before padding/overlay. Furthermore, there are no constraints (curb, gutter) to warrant milling before overlay; this strategy may prove valuable if these constraints exist or if the road was showing significant distress but had sufficient strength and the intent was to retard reflection cracking. Table 5 illustrates this recommendation, with consideration for uniform thickness across the road and for reasonable lengths. Routine maintenance should be applied on all portions not receiving an overlay. The overlay work should be carried out within the next two years; otherwise the surface distress and deflection data might be invalid.

The case illustration of the foregoing paragraphs involves two key points:

1. Automation of the deflection information in terms of computer graphics plotting, and calculation of overlay thicknesses corresponding to the deflection, as in Figure 6 relieves the designer from a lot of drudgery and allows him/her to actually do more in-depth analysis and creative interpretation as illustrated in Table 5. In other words, all of the front end work can be automated and then the designer can look at a Figure 6 type plot, plus the surface distress report of Table 5 and other information, and quickly arrive at a final overlay recommendation for the entire road width.

2. Deflection can vary quite markedly along the road length, and between lanes, as shown by the deflection profiles of Figure 6. This suggests that it is better to capture such variation by using the type of overlay design approach shown in Figure 6 and Table 5, as compared to using a more complex method but with only average data for a section. In fact, with the type of balanced final design of Table 5, which involves practical variations of thickness along the section length, it is probable that there would actually be savings in material quantities. Moreover, because of the varying thicknesses, the deflection profile on the overlayed pavement should be much more uniform.

TABLE #4

FINAL REHABILITATION RECOMMENDATIONS FOR

TAYLOR BOULEVARD

(from Mona Drive to Gardiners Road)

STATION (meters)	RECOMMENDATION*
0+000 – 0+225	No overlay
0+225 – 0+400	100 mm of padding + 50 mm overlay
0+400 – 0+550	50 mm overlay
0+550 – 0+750	50 mm of padding + 50 mm overlay
0+750 – 1+250	No overlay
1+250 – 1+800	50 mm overlay
1+800 – 1+950	100 mm of padding + 50 mm overlay
1+950 – 2+150	50 mm overlay
2+150 – 2+300	50 mm of padding + 50 mm overlay
2+300 – 2+825	No overlay

(*) Any severely alligatored portions on a localized basis should be dug out and replaced prior to padding or overlay and severe cracks (longitudinal and transverse) should be filled prior to padding or overlay.

CONCLUSIONS

Thin asphalt pavements may require overlays for several reasons but inadequate structural capacity is usually a major one. Deflection testing and analysis provides a sound basis for determining the thickness of overlay required.

However, the way in which such deflection test results can be used ranges from the empirical, where sufficient overlay thickness is provided to reduce the surface deflection to some maximum tolerable value, to the theoretical or mechanistic where materials properties and elastic layer models-are used. The latter is certainly a desirable approach but the current state of technology makes its applicability to thin asphalt pavements somewhat limited. Consequently, the empirical method is, for the time being, still more appropriate for thin asphalt pavements.

Whatever method is used, though, it is highly desirable that the layer variations in deflection along a section length and between lanes be adequately reflected in the overlay design. A convenient way of doing this is to automate the process to the point where the designer can look at ideal thickness variations along the section for each lane, plus other information on surface distress, etc., and arrive at a balanced, practical overlay design.

REFERENCES

1. Roads and Transportation Association of Canada, “Pavement Management Guide”, 1977.

2. Haas, Ralph and W. Ronald Hudson, “Pavement Management Systems”, Krieger Publishing Co., 1978.

3. Karan, M.A., Ralph Haas, D.A. Kobi and A. Cheetham, “Implementation and Verification Examples of Successful Pavement Management”, Proc., Int. Conf. on Structural Design of Asphalt Pavements, Univ. of Michigan, 1982.

4. Phang, W.A. and Ralph Haas, “Draft Report on National Workshops on Pavement Management Resource Requirements Task Group”, Prepared for TBB and FHWA, December, 1980.

5. Eaglesen, B., Heisey, S., Hudson, W.R., Meyer, A.H. and Stokoe, K., “Comparison of the Falling Weight Deflectometer and the Dynaflect for Pavement Evalution”, RR 256-1, Center for Transp. Research, University of Texas at Austin, November 1982.

6. Bush III, A.J., “Non-Destructive Testing In Light Aircraft Pavements – Phase I, “U.S. Army Waterways Experimental Station, Vicksburg, Mississippi, August 1979.

7. Swift, “A Graphical Technique for Determining the Elastic Moduli of a Two-Layered Structure from Measured Surface Deflection”, HRB Record 431, 1973.

8. Vaswani, N.K., 1971. Method for Separately Evaluating Structural Performance of Subgrades and Overlaying Flexible Pavements, Record 362, Highway Res. Bd., pp. 48-62.

9. Vaswani, N.K., 1974. Estimating Moduli of Pavement Materials from Field Deflection Measurements, Report VHTRC 75-R20, Virginia Highway and Transp. Res. Council, Charlottesville, VA. 29 pp.

10. Vaswani, N.K., 1977. Determining Moduli of Materials from Deflections, Trans. Engrg. Jour., ASCE, 103 (TE1): 125-141.

11. Wiseman, G., 1973. Flexible Pavement Evaluation Using Hertz Theory, Transp. Engrg. Jour. ASCE, 99(TE 3): pp. 449-466.

12. Wideman, G., et al. 1977. Simple Elastic Models for Pavement Evaluation Using Measured Surface Deflection Bowls, Proceedings, 4th International Conf. on the Structural Design of Asphalt Pavements, Ann Arbor, MI.

13. U1lidtz, P., 1973. The Use of Dynamic Plate Loading Tests in Design of Overlays, Report 22, Technical University of Denmark, Copenhagen, 26 pp.

14. Ullidtz, P., 1977. Overlay and Stage by Stage Design, Proceedings, 4th Inter-national Conf. on the Structural Design of Asphalt Pavements. Ann Arbor, MI. pp. 722-735.

15. Irwin, L.H., “Equipment and Methods for Deflection-Based Structural Evaluation of Pavements”. Proceedings – Forest Service Geotechnical Workshop – Ames, Iowa, 1979.

16. Irwin, L.H., and Johnson, T.C., “Frost-Affected Resilient Moduli Evaluated With the Aid of Nondestructive Measured Pavement Surface Deflections”. Paper presented to TRB-A2T56 Workshop on “Nondestructive Evaluation of Airfield Pavements”, August 1981.

17. Hoffman, M.S., Thompson, M.R., “Mechanistic Interpretation of Nondestructive Pavement Testing Deflection, FHWA/IL/UI-190, June 1981.

18. Hoffman, M.S., Thompson, M.R., “Nondestructive Testing of Flexible Pavement Field Testing Program Summary”, FHWA/IL/UI-188, June 1981.

19. Thompson, M.R., “ILLI-PAVE Procedure for Determining In-situ Properties from NDT Deflection Data”. Paper presented to TRB-A2T56 Workshop on Nondestructive Evaluation of Airfield Pavements, August 1981.

20. Bush, J.J., “Pavement Evaluation Using Deflection Basin Measurements to Characterize an Elastic Layer System”. Paper presented to TRB-A2T56 Workshop on Nondestructive Evaluation of Airfield Pavements, August 1981.

21. Shahin, M.Y., Sharma, J., Smith, R.E., Stubstad, R.N., “Nondestructive Testing of Airfield Pavement – The Dynplot/ERES Method”. Paper presented to TRB-A2T56 Workshop on the Nondestructive Evaluation of Airfield Pavements, August 1981.

22. Larsen, H.J., Stubstad, R.N., “Use of Nondestructive Testing in Flexible Pavement Rehabilitation Design”. Proceedings Vol. 1, Bearing Capacity of Roads and Airfields, Norweigan Institute of Technology, Trandheim, Norway, 1982.

23. Ullidtz, P., “Overlay and Stage-by-Stage Design”. Proceedings Fourth Inter-national Conference on the Structural Design of Asphalt Pavements, Ann Arbor Michigan, 1977.

24. Koole, R.C., and Visser, W., “Aircraft Pavements, Evaluation and Overlay Design”. Proceedings, Volume 2 Bearing Capacity of Roads and Airfields, Norweigan Institute of Technology, Trandheim, Norway, 1982.

25. Wong, M.C., “Pavement Layer Modules Evaluation From Deflection Basin”. Pro-ceedings Volume 2 – Bearing Capacity of Roads and Airfields, Norweigan Institute of Technology, Trandheim, Norway, 1982.

26. Kilareski, W.P. and Bassan, A.A., “Evaluation of In-situ Moduli and Pavement Life from Deflection Basins”. Proceedings, Fifth International Conference on the Structural Design of Asphalt Pavements, Delft University, Netherlands, 1982.

27. Transportation Research Record 666, “Pavement Surface Properties, Evaluations, and Shoulders”, TRB, 1978.

28. Majidzadeh, K., Ilves, G., McComb, R., “Overlay Design of Flexible Pavements Using Dynaflect”. Proceedings Volume Z Bearing Capacity of Roads and Airfields, Norweigan Institute of Technology, Norway, 1982.

29. Sharpe, G.W., Southgate, H.F. and Deen, R.C., “Pavement Evaluation Using Dynamic Deflections”, TRB Research Record 700, 1979.

30. Southgate, H.F., Sharpe, G.W., Deen, R.C., and Havens, J.H., “Structural Capacity of In-Place Asphaltic Concrete Pavements from Dynamic Deflections”. Volume l, Fifth International Conf. on Structural Design of Asphalt Pavement Delft University, Netherlands, 1982.

31. Epps, Jon A. and R.G. Hicks, “Moderators Summary of Session V on Rehabilitation of Pavements”, Proc., Vol. II, Fifth Int. Conf. on the Structural Design of Asphalt Pavements, Univ. of Michigan, 1982.

32. Asphalt Institute MS-17 – Asphalt Overlap and Pavement Rehabilitation College Park, Maryland, 1969.

33. Lister, N.W., C.K. Kennedy and B.W. Berne, “The TRRL Method for Planning and Design of Structural Maintenance”, Proc., Vol. I, Fifth Int. Conf. on the Structural Design of Asphalt Pavements, Univ. of Michigan, 1982.