Measurement of pavement deflections in tropical and sub-tropical climates by H. R. Smith and C. R. Jones F Transport and Road Research Laboratory TECHNICAL INFORMATION AND LIBR,ARY SERVICES Please return this document to the Librarv, or request an extension of loan, by the date Iast stamped below ACCESSION NUMBER —— ___ S303 TRANSPORT and ROAD RESEARCH LABORATORY Department of the Environment Depatiment of Transport TRRL LABORATORY REPORT 935 MEASUREMENT OF PAVEMENT DEFLECTIONS INTROPICALAND SUB-TROPICAL CLIMATES by HRSmithand C RJones . The work described in this Report forms part of the programme ~rried out for the Overseas Development Administration, but any views expressed are not necessarily those of the Administration Overseas Unit Transport and Road Research Laboratory Crowthorne, Berkshire 1980 ISSN 0305–1 293 , CONTENTS Abstract 1. Introduction :2. The deflection beam 2.1 Details of the deflection beam 2.2 C~bration of the deflection beam 2.3 Theloaded lorry 2.3.1 Thespecification adoptedby 2.3.2 Other specifications -- the OverseasUnit 2.4 Methods ofustig thedeflection beam 2.4.1 Thetransient deflection test 2.4.2 The rebound deflection test 3. Factors affecting deflection measurements 3.1 hfluence of loading wheels upon the deflection beam 3.2 Effect of changes in solar radiation on the deflection beam 3.3 The condition of the road surface 3.4 The effect of temperature 3.4.1 .The effect of plastic flow upon the deflection test method 3.5 Effect of hardening of the bitumen in a road pavement 3.6 Effect of seasonal rainfall 4. hcfllary equipment 4.1 Temperature measurement 4.2 Pavement condition 4.2.1 Deformation 4.2.2 Cracking 5. Measurement of radius of curvature Page 1 ‘1 2 2 2 3 3 3 4 4 5 5 5 6 6 7 8 8 8 9 9 9 9 9 10 Page 6. Conclusions 11 7. Acknowledgements 11 8. References 11 9. Appendti: A deflection survey method for evaluating roads in developing countries 21 9.1 htroduction 21 9.2 ~flection survey methods 22 ‘- Q CROWN COPYRIGHT 1980 Extracts from the text may be reproduced, except for commercial purposes, provided the source is acknowledged MEASUREMENT OF PAVEMENT DEFLECTIONS IN TROPICAL ANDSUB-TROPICAL CLIMATES WSTRACT The measurement of the deflection of the surface of road pavements under astandard wheel load iscommon practice in many countries. The main purposes of making deflection measurements are to assistin the evaluation of the structural condition of roads, and to help in the design of road strengthening measures. h developing countries in the tropics the high road temperatures and other climatic effects necessitate the use of a special technique in making deflection tests. This report describes the techniques adopted by the OverseasUnit of the Transport and Road Research bboratory for the measurement of pavement deflection in tropical developing countries, and discussesfactors relevant to the deflection of road pavements in such conditions. In many countries systematic .A - 1. INTRODUCTION measurements are made of the deflection of road surfaces tinder a standard wheel load 1YZ’5.These deflection surveys are made for two main reasons:– i) to help in estimating the ability of existing roads to carry future traffic loads and to predict the optimum time for strengthening them, ii) to assist with the design of the thickness of strengthening overlays for road pavements. hr developing countries these uses of deflection surveys are particularly relevant because many of the main roads were origina~y built to carry relatively light traffic, and now they urgently require strengthenkg to carry the much heavier,traffic of the present day. It is usudy more cost+ ffective to strengthen a road than to reconstruct it, particularly if the thickness of the overlay is varied to match variations in the strength of the existing pavement. Deflection surveys enable stlengthening overlays to be designed to match the existing pavement, or indicate whether reconstruction is required4*5. Detafls of the standard method adopted for measuring pavement deflections in the United Kngdom ‘have been published 67Y. In the tropics, high road temperatures and other cfimatic effects necessitate the use of somewhat different deflection test procedures. This report describes the deflection technique used in tropical and sub-tropical developing countries by the OverseasUnit of the Transport and Road Research bboratory, and discussesfactors relevant to the deflection of road pavements in such conditions. A method of making deflection surveys for assesskg overlay thickness requirements in developing countries has been recommended earfier8, a revision of this method is included in the Appendix of this report. 1 2. THE DEFLECTION BEAM 2.1 Details of the defletiion beam Deflection measurements are normally made with a beam of the type designed by A C Benkehnang, used in conjunction with a loaded lorry. The beam has a long slender pivoted arm which can be placed between the dud rear wheels of a lorry. The pivot is one-tfird of the way along the beam and is carried by a frame which is supported on three adjustable legs. A dial gauge is mounted on the frame in such a way that it measures movement of the toe of the beam resting on the road surface. The design of the beam used by the Transport and Road Research hboratory is illustrated in Hate 1 and the principle dimensions of the beam are given in Figure 1. A narrow sunshade is fitted as standard to the beam to reduce distortion induced by changes in radiation fting on it. For research purposes in the tropics a special sunshade which covers the frame and that part of the beam which is not placed between the lorry wheels is used. This practicably eliminates distortion of the instrument due to changes in radiation fafling on it. The dial gauge has a 75 mm diameter face, a travel of 25 mm and 0.01 mm graduations with numbers printed in reverse to enable them to read from above using a mirror mounted at 45° to the gauge. A locking device protects the did gauge w~st the beam is being moved and is also used to lock the beam to its frame for transportation. An electric ‘buzzer’ with a battery and hand switch are mounted on the frame to vibrate the beam pivot and dial gauge to ensure free movement during testing. The beam can be split into two parts for ease of transportation and it is therefore essential that the connecting plates are tightly secured before use. It is necessary to treat the dial gauge with care; the stem of the gauge should be polished with a dry, soft cloth and checked for free movement before testing is started. The stem should dso be polished periodically during each day of testing to remove any deposits. 2.2 Calibration of the defletiion beam The dial gauge and beam should be calibrated before use to ensure that they are working correctly. A calibration rig, which can be suppfied by the beam manufacturer, is shown in Plate 2. The rig is fitted - with a variable cam which can impart vertical movements of different magnitudes to the toe of the beam, simulating deflections. With the vibrator working several readings on the dial gauge on both the beam and calibration rig are recorded for each cam setting. If the dial gauge on the beam does not read one half of the dial gauge reading on the ctibration rig, the tightness of securing bolts should be examined and the dial gaugesshould be checked and replaced if necessary. If the gauges are functioning correctly, the beam pivot should be examined and checked for free and smooth operation and the striking plate beneath the dial gauge spindle should be checked to ensure that it is tightly secured and has not become grooved by the did gauge stylus. The amount of vibration produced by the buzzer must be such that only sfight movement of the dial gauge needle can be detected when the buzzer is actuated. 2 2.3 The loaded lorry . 2.3.1 The specification adopted by the Overseas Unit. For the deflection method used by the OverseasUnit of the Transport and Road Research bboratory a two axle lorry, with twin wheel sets on each end of the rear axle, is loaded so that the load on the rear de is 62.3 kN (6350 kg~, equafly divided between the twin wheel assemblies. Blocks of dense concrete or boxed scrap metal are ideal materials for loading the lorry. It is important that the load should not be able to move, nor should it be capable of absorbing or trapptig water which would cause the magnitude of the axle load to change. The general dimensions and weights of the vehicle given in Table 1 are recommended. The use of cross-ply tyres rather than radial tyres is recommended because it is norma~y easier,to obtain the correct gap between the wds of the tyres on the dud wheels with cross-ply tyres. If tyres other than those recommended are used then the tyre pressures may have to be adjusted to achieve contact areas as indicated in Table 1 and Figure 2. TABLE 1 Details of vehicles suitable for deflection testing Characteristic Rear ade load Dual wheel load Front axle load Wheel base Tyre size Tyre pressure Gap between walls of dud rear tyres Gap between contact area of dud rear wheels Contact area of twin rear wheels Specification 62.3 kN (6350 kg~ * 5% 31.1 kN(3175kg~* 5% 22.6–32.4 kN (2300–3300 kg~ 3.85m (approx) 7.50 x 20) 8.25 X 20) ‘referred 9.00 x 20 acceptable 590 kN/m2 (85 psi) 25–40 mm ~ 30 mm is recommended) 100–1 50 mm Simflar to that shown in Figure 2 2.3.2 Other specifications. A common rdternative to the specifications given in Table 1 involves the use of a vehicle withan80kN(8165 kg~ rear axle load, with consequently larger tyres. Overlay design methods developed on the basis of the heavier ade load are equdy as valid as those developed on the basis of a 62.3 kN tie load. Care must be taken however not to confuse the deflection criteria applicable to one tie load with those applicable to the other. The advantages of using a 62.3 kN rear tie load are that a sma~er lorry can be used and running costs til be generdy lower. With few exceptions, the magnitudes of the deflections measured in the two cases are directly proportional to the wheel loads used. 3 2.4 Methods of using the defledion beam There are two basic methods which are commonly used for operating the deflection beam. 2.4.1 Tha transient defledion test. This type of test has been adopted for measuring deflections in the United ~gdom and is the method used by the OverseasUnit of the Transport and Road Research hboratory in developing countries. The procedure is to mark the test point in the verge side wheelpath and to position the lorry par~el to the road edge with its rear ade 1.3m behind the test point. The lorry is aligned such that when it moves forward the test point bisects the distance between the tyres of the verge side dud wheels. With the lorry in the initial stationary position a deflection beam is positioned centrally between the dual rear wheels with its measuring toe resting on the test point, 1.3m in front of the rear axle. By using the adjustable legs and the ‘buflt-in’spirit levelthe frame of the beam is levelled transversely and having checked the alignment of the beam and adjusted it if necessary, the locking device on the beam is released. Adjustment of the rear foot wifl ensure that the free end of the sun-shade is clear of the beam and that there is adequate travel of the dial gauge spindle to record the deflection. A suitable method of judging this is to adjust the rear foot so that when the end of the beam under the dial gauge is pushed upwards manuafly (flexing the beam in the process), 4 to 6 revolutions on the dial gauge are registered before travel ceases. The vibrator is switched on and the dial gauge scale is rotated until a reading of zero is indicated. When the beam operator has completed his preparations the lorry can be driven forward at creep speed to a point such that the rear axle is not lessthan 5m beyond the test point. The time taken for the lorry to move a distance of 5m from the starting position should be between 9 and 11 seconds, this time should be checked at regular intervals during each day. Sometimes during the period between setting the dial gauge to zero and the lorry actu~y starting to move, the dial gauge reading may change slightly. It is recommended that this new reading is recorded asthe initird reading rather than resetting the dial gauge to zero. Records are kept of the initial, maximum and final dial gauge readings. The difference between the initial and maximum readings is proportional to the transient loading deflection and the difference between the maximum and final readings is proportional to the transient recovery deflection. Because the lengths of the arms of the pivoted beam are in the ratio of 2:1 the differences between the dial gauge readings must be doubled to obtain the actual deflection and recovery of the road surface. The transient deflection is the mean of the loading and recovery deflections in the transient test. The principal positions of the lorry during the test are shown in Plate 3. h order to measure transient deflections quic~y with ne@gible risk of contact between the tyres and the deflection beam, adjustable pointers should be fitted to the lorry chassisin front of both dual wheel sets as shown in Plate 4. They must be adjustable so that they can be afigned to point to the road surface approximately 1.3m in front of the rear afle and directly in fine with the path taken by the centreline of the dud wheel set. Correct fignment of the pointer can be obtained in the following way. Position the lorry as for a deflection test, tign a suitable straightedge (not the deflection beam) so that it lies centrdy between the dud wheels with one end 1.3m in front of the rear axle and adjust the pointer so that it points to the end of the straight edge. By moving the lorry forward in one metre increments the pointer alignment can be adjusted untfl it is correct. 4 Two tests are normrdly carried out on each test point and the difference between the tests should be within the fimits recommended in Table 2. If the readings differ by more than the limits given then additional tests should be carried out until acceptable repeatabdity is obtained. If a characteristic deflection for the test point is not obtained after 5 tests have been carried out, then the best estimate from these readings should be taken. TABLE 2 Repeatability of duplicate deflection tests Mean deflection Maximum permissible difference (x 10-2 mm) between the two tests (x 10-2 mm) <10 2 10– 30 3 31– 50 4 51–100 5 >100 6 It should be remembered that due to friction the beam tends to register vrdueswhich are lower than the true deflections, and adequate vibration and correct operation of the beam pivot and dial gauge are essential. It is common practice to measure deflections simultaneously in both wheelpaths, but when the lorry is positioned to measure the deflection on the centreline of the verge side wheelpath it may be found that the offside wheels are not in the centre of the offside wheelpath. Additional tests wfllindicate if the error resdting from measuring the deflection to one side of the centrehne of the wheelpath is large enough to warrant the repositioning of the lorry before testing in the offside wheelpath. 2.4.2 The rebound defle~ion test. The rebound deflection test differs from the previous test in that the dud wheels,are positioned immediately above the test point and the toe of the beam is placed on the test point between the dual wheels. The beam is adjusted in the same way as for the transient test and when the initial reading has been noted, the lorry is driven forward at creep speed untfl the wheels are far enough away to have no influence upon the deflection beam. The find dial gauge reading is recorded and the ‘rebound deflection’ is twice the difference between the initial and final dial gauge readings. Because the rebound deflection can be influenced by the length of time during which the loading wheels are stationary over the test point care must be taken over the exact procedure used3. The rebound test is not recommended by the OverseasUnit of the Transport and Road Research hborat ory, for reasons which are discussed in detd in Section 3.4.1 of this report. 3. FACTORS AFFECTING DEFLECTION MEASUREMENTS 3.1 InfIuen@ of loading wheels upon the defledion beam At the beginning of the transient deflection test the loading wheels are 1.3mfrom the toe of the beam and 1.44m from the twin feet of the beam. Occasiona~y when the lorry is in this position the toe and the !j front feet of the beam are within the deflected bowl of the road surface under the loaded wheels. In rare instances small displacements of the rear foot of the deflection beam have also been detected. On roads with crushed stone or weak soil-cement road bases overlying subgrades with CBRvalues in excess of 8 per cent, the error in the dial gauge reading caused by the influence of the loading wheels, when the lorry is in 10. the initial postiion can be calculated from the following equation . d=~D_lst~K where, 5 10 d = error in dial gauge reading D= depression of the toe of the beam s = depression of the front feet of the beam K= depression of the back foot of the beam, The error in the measured transient deflection will be of the order of 0.9 x (the movement of the front feet during the test) or less depending upon the exact shape of the deflected bowl and the measured deflection will, on the roads described above, be larger than the actual deflection. This solution wifl not necessarily apply where a stiff pavement has been constructed on a weak subgrade or on roads with very stiff soil-cement road bases. An example of how a second deflection beam can be used to measure the movement of the beam feet during a deflection testis illustrated in Plate 5. The influence of the front and rear wheels of the lorry on the toe of the beam can be investigated by varying the position of the lorry and beam. Movements of the feet of deflection beams recorded during extensive deflection surveys in Kenya were only occasionally greater than 6 x 10-2 mm. It is recommended that corrections for movements of this magnitude are not made, and that the use of non-absolute deflections is accepted. If the residual reading on the dial gauge indicates that the magnitude of the movement of the feet of the beam is larger than this, a correction can be made using a second deflection beam as shown in Plate 5. 3.2 Effeti of changes in solar radiation on the defledion beam The deflection beam is sensitive to changes in radiation which can occur in tropical environments when cloud shadow moves onto or away from the beam. A change in dial gauge reading of * 4 x 10-2 mm, depending upon whether the shadow moves onto or away from tie beam, can occur within ten seconds. me tip of the beam, which is shaded by the lorry at the start of a transient deflection test, is less sensitive to changes in radiation than the reference frame and the rear end of the beam. For research purposes the OverseasUnit has fitted doy ‘sunshades’to its deflection beams (see Plate 1). Each section of the sunshade is free to move at one end. The beams are then virtually insensitive to changes in radiation. If deflection tests are carried out during continuously cloudy or sunny conditions, errors due to changes in radiation will not occur. 3.3 The condition of the road surfaw Consideration must be given to the condition of the road surface, particularly on older road pavements. Badly crazed surfacings may be loose and can cause considerable errors in deflection measurements. It is 6 often possible to remove loose material and measure the deflection on a firmer lower layer. In this case it may be necessary to have avaflable a beam with an adjustable toe which can reach down to the firmer surface. Errors can also be caused if the surfacing layer is not bonded to the underlying structure. During hot weather the feet of the beam may sink into the soft bitumen on roads with rich surface dressings. On coarse textured surfaces the positions of the toe and feet of the beam must be selected so that they remain stable during the test. 3.4 The effeti of temperature The stiffness of bituminous materials is susceptible to changes in temperature 711‘ and therefore the pavement deflection can also be affected by temperature changes. The magnitude of this effect is dependent upon the proportion of the overafl pavement stiffness provided by the bituminous material. In certain cases the te-mperature effect can be such that measured deflections are meanin~ess, particularly on new or bitumen. rich surfacing layers. h undertaking deflection surveys it is practical to measure the pavement temperature only on the surface or near the surface and it is recommended that temperature measurement at a standard depth of 40 mm is adopted. In tropical environments there are normally many days during the year when road temperatures ~ncreasecontinuously during the morning and early afternoon followed by a period when the surface temperature may remain constant or begin to decrease; at this stage the temperature gradient in the pavement is significantly different from when the temperature of the whole pavement is rising. mere the surfacing material is sensitive to changes in temperature there are four main phases in the deflectiontemperature relationship. The first phase occurs early in the morning before the road temperature begins to increase. At these lower temperatures some surfacings are considerably stiffer than at higher temperatures, resulting in much smafler deflections being measured. It is sometimes very difficult to obtain rehable :relationships between temperature and deflection for these conditions. During the second phase as the road temperature rises a clear relationship may be obtained between increasing temperature and deflection. Depending upon the nature of the surfacing material a third phase can occur as the temperature rises :further in which there is an increasing tendency for the surfacing material to flow plastica~y upwards between the dud load wheels of the deflection lorry as the temperature of the surfacing increases. The plastic flow can be so large that the measured deflections are meanin~ess. Findy the pavement deflection may continue to increase without an increase in surface temperature, because the temperature gradient in the pavement continues to change as lower layers warm up. It is not possible to give general guidance on the magnitude of temperature-deflection corrections for roads in tropical and sub-tropical environments 12 since such corrections are ve~ much dependent on the type of road pavement. Roads with thin sealssuch as surface dressings are usu~y unaffected by changes in temperature, but other roads with substantial hyers of bituminous materird incorporated in their construction must be tested to detertie their temperature-deflection characteristics. For tropical ~d sub-tropical conditions the Overseas Unit of the Transport and Road Research hboratory has adopted a standard reference temperature of 35°C measured at a depth of 40 mm, to which deflections measured on temperature-susceptible materials are corrected. 7 3.4.1 The effed of plastic flow upon the defletiion test method. men the surfacing material flows plastica~y it squeezes upwards between the durd loading wheels of the deflection lorry 12 which, in the transient deflection test, reduces the transient loading deflection because the upward movement of the material count eracts the downward movement of the pavement. The transient recovery deflection that is . measured may be correct but further plastic movement of the raised surfacing material can occur during the time tiken for the wheels to move from the test point to the final position, thereby causing an error in the recovery deflection reading. One of the reasons why the transient deflection testis preferred for use in the tropics is that it is very clear from the test results when plastic flow occurs. Provided that a deflection survey team adopts a standard procedure, an engineer who may not be on site can tell from the field record sheets when plastic flow has occurred by comparing the loading and recovery deflections from the transient test. An example of plastic flow occurring during temperature-deflection tests is shown in.Figure 3. . h the rebound test with the wheels located directly over the test point at the start of the deflection test, greater plastic flow wi~ be induced in susceptible materials because of the time during which the wheels remain stationary in this position. men the lorry is driven forward the road surface ‘rebounds’ but an indeterminate amount of recovery of the displaced surfacing material can occur. There is thus no clear indication from the simple rebound test when plastic flow occurs. 3.5 Effe~ of hardening of the bitumen in a road pavement It has been found that the bitumen in surfacing materials can harden very rapidy in tropical environments 2’13. A nominal 80/100 penetration grade bitumen recovered from dense mixes can have a penetration vrdue14 of 20 after four years in the road surface. The effect of this hardentig is to-stiffen o the bituminous layer. An example is shown in Figure 3 where the plastic flow of a bituminous surfacing material at a given temperature decreases with the ageing, or hardening, of the bitumen. 3.6 Effed of seasonal rainfall Periods of rain or dry weather can cause changes in moisture conditions under pavements, thereby affecting the overall stiffness of the structure and the deflection measured on it. An example of the effect of seasonal rainfall upon deflections is given in Figure 4. It is important to measure pavement deflections at the end of the wettest period of the year, since this is when the pavement is weakest. In many regions annual rainfall patterns vary considerably from year to year, in which case it is desirable to study the longer term variations in rainfa~ and deflection that occur in a region to assistin interpreting deflection survey results obtained in the region. mere long periods of drought occur itis important that dowance is made for the increase in pavement deflection which can occur during wet periods, a factor that is particularly important when the thickness of bituminous overlays is being decided upon. 8 @ 4. ANCILLARY EQUIPMENT 4.1 Temperature measurement The temperature within bituminous layers can have a significant effect upon deflections, therefore frequent measurements of temperature must be made during surveys on pavements which contain bituminous materials. The exact procedure adopted wifl depend upon the results of temperature-deflection relationships determined for the particular road. The method of determining temperature-deflection relationships used by the OverseasUnit is to measure repeat edly the deflection of one point on the road as the temperature increases during a day. Temperatures are measured at depths of 40 mm and at 75 mm and 110 mm if bituminous byers of more than 75 mm and 100 mm in thickness have been used. A number of adjacent points can be tested during thi same day, the number being such that approximately ten tests can be completed on each test point during the day. A limited number of tests are carried out as the surfacing temperature decreases. It has been found that for deflection surveys adequate corrections for temperature effects can be based upon the measurement of temperature at a depth of 40 mm, but it is advisable to check this on pavements which have thick bituminous layers. During deflection surveys carried out on temperature-sensitive materials, the temperature at a depth of 40 mm should be measured adjacent to each test point. A simple method of measuring the temperature is to make a hole with a 7 mm diameter percussion drfllto such a depth that the centre of the bulb of a mercury in ~ass thermometer is at a depth of 40 mm. Enough #ycerol or oflis placed in the hole so that the thermometer bulb is covered. The thermometer should preferably be one with a short bulb and which requires a sma~ depth of immersion. 4.2 Pavement rendition Deflection values are most useful when they are accompanied by other indicators of pavement performance. Measurements of rut depth, cracking in the wheelpaths, and crack depth, are simple measures of pavement condition that can easily be made at the same time that deflection surveys are being carried out. 4.2.1 Deformation. Deformation or rut depth in the wheelpaths can be measured by placing a 2rn straight edge transversely to the road over the test point, any deformation below the straight edge being measured with a cdbrat ed wedge. 4.2.2 Cracking. An assessment of the amount of cracking at a deflection test point can be made by measuring the hear cracking within a one metre square frame. The method developed by the Overseas Unit is to place a one metre square frame over the deflection test point with one side parallel to the edge of the road such that it encloses the maximum length of cracMng present, provided that the test point is within the frame. The frame is not placed across the centre line of the road, but centre line cracks are recorded separately since these may be a result of construction faults rather than the effects of traffic. The cracks can be ‘chaked h’ and measured with a flexible tape. @ It has been found that cracking in overlay trials constructed in Kenya invariably starts at the surface. The development of these cracks and the seriousness of them can be assessedfrom cores cut ‘down through the cracks. A convenient means of classifying the surface condition at a test point is shown in Table 3. TABLE 3 Classification of road surface condition Tr~sverse deformation under a 2m long Degree of cracking (visible cracks) straight edge ~assification Index Deformation Classification Index Crack length/unit area D, I bssthan 10 mm I C, I NIL D2 IOmmto 14mm C2 Not greater than 1 m/m2 D3 15mmto19mm C3 Greater than 1 m/m2 but not greater than 2 m/m2 D4 20 mm to 25 mm C4 Greater than 2 m/m2 but not greater than 5 m/m2 Greater than 5 m/m2 D5 Greater than 25 mm C5 (ravening & potholing imminent, immediate maintenance required) 5. MEASUREMENT OF RADIUS OF CURVATURE It is generally accepted that the measurement of deflection together with the longitudinal radius of curvature of the road surface enables a better estimate to be made of the structural condition of an existing pavement than the measurement of deflection alone. The combined measurements should thus provide a better correlation with overlay performance, particularly where pavements of different or widely variable construction are overlaid15~16. There are a number of ways of measuring the radius of curvature of the surface of a deflected pavement 16’17. The method adopted by the Ove~seasUnit of TRRL is to use a deflection beam fitted with an LVDT in place of the dial gauge. The LVDT is connected to a chart recorder, the paper of which is driven by a pulse generator actuated by the forward movement of the deflection test lorry. The power source is a pair of 12 volt lead acid batteries. This system provides an influence line obtained during the transient deflection test which represents the shape of the deflected road surface under the loaded dud wheels of the lorry. The magnitude of the transient deflection and an estimate of the radius of curvature can be read from the chart. The radius of curvature is estimated from the difference between the maximum deflection and the deflection measured when the loading wheels are at a point 125 mm from the point of maximum deflection. \ This is simflar to the method proposed by DeMenl 8. 10 kdius of curvature = ~ 2d where L = 125 mm d = ‘differential deflection A simpler way of measuring the deflected shape of the road surface is to carry out a normal deflection test, except that the test lorry is moved forward in a series of increments, the dial gauge being read at each increment. An influence line can then be plotted. The disadvantages of this method are that it is very slow and that creep may occur within the pavement layers under the stationary loaded wheels which may significantly affect the magnitudes of the deflection readings. Important advantages are that the test is simple and can be performed with the standard deflection beam. 6. CONCLUSIONS When measuring pavement deflections in tropical ctimates using deflection beams and a standard procedure, it is necessary to take account of a number of factors which can influence the magnitude of the recorded deflection. These are:– i) the influence of the loaded wheels on the support frame of the deflection beam, ii) the effect of solar radiation on the beam, iii) the condition of the road surface, iv) the effect of temperature on bituminous materials, v) the effect of the hardening of bitumen in the bituminous layers of a pavement, vi) the effect of seasonal rainfa~ on pavement strength. 7. ACKNOWLEDGEMENTS The work described in this report was carried out in the OverseasUnit (Head of Unit: Mr J N Bulman) of TRRL. 8. REFERENCES 1. HIGHWAYRESEARCH BOARD. Design of overlays and pavement rehabilitation. Highway Research Record 300. Washington, 1969. 2. HIGHWAYRESEARCH BOARD. Evaluation of pavements by deflection studies for maintenance purposes. Highway Research Record 129. Washington, 1966. 3. CANADIAN GOOD ROADS ASSOCIATION. A guide to the structural design of flexible and rigid pavements in Canada. Canadian Good Roads Association Pavement Design and Evaluation Committee. Ottawa, 1965 (Canadian Good Roads Association). 4. KENNEDY, C K and N W LISTER. Prediction of pavement performance and the design of overlays. Department of the Environment Department of Transport, TML Report LR 833. Crowthorne, 1978 (Transport and Road Research Laboratory). 11 . 5. ORGANISATION FOR ECONOMICCO-OPERATION AND’DEVELOPMENT. Road Strengthening. A report prepared by an OECD road research group. Paris, 1976 (Organisation for Economic Cooperation and Development). 6. KENNEDY, C K, P FEVRE and C CLARKE. Pavement deflection: equipment for measurement in the United Kingdom. Department of the Environment Department of Transport, T~L Report LR 834. Crowthorne, 1978 (Transport and Road Research Laboratory). 7. KENNEDY, C K. Pavement deflection: operating procedures for use in the United Kingdom. Department of the Environment Department of Transport, TUL Report LR 835. Crowthorne, 1978 (Transport and Road Research Laboratory). 8. SMITH, HR. A deflection survey technique for pavement evaluation in developing countries. Department of the Environment, TUL Report LR 525. Crowthorne, 1973 (Transport and Road Research bboratory). 9. HIGHWAYRESEARCH BOARD. The WASHORoad Test. Part 1. Design, construction and testing procedures. Highway Research Board Special Report No 18. Washington, 1953. 10. LISTER, N W. A deflection beam for investigating the behaviour of pavements under load. Road Research Laboratory Research Note RN/3842/NW2, 1960 (Unpublished). 11. CRONEY, D and J N BULMAN. The influence of climatic factors on the structural design of flexible pavements. Proc. Third Int. Con& on the StructuralDesi@ of Asphalt Pavements. hndon, 1972. Ann Arbor, 1972 (University of Michigan). 12. JONES, C R and H R SMITH. Deflection-temperature relationships for bituminous road surfacings in Kenya. Department of the Environment Department of Transport, T~L Report LR 936. .Crowthorne, 1980 (Transport and Road Research Laboratory). 13. BULMAN,J N and H R SMITH. Pavement performance and deflection studies on Malaysian roads. Department of the Environment Department of Transport, TML Report LR 795. Crowthome, 1977 (Transport and Road Research bboratory). ~ 14. INSTITUTE OF PETROLEUM. 1PStandards for petroleum and its products. Part 1: Methods of analysis and testing. London, 1978 (Institute of Petroleum). 37th Edition. 15. DEHLEN, G L. Flexure of a road surfacing, its relation to fatigue cracking and factors determining its severity. H~ Bulletin 321, 1962. 16. LEGER Ph and P AUTRET. The use of deflection measurements for the structural design and supervision of pavements. Proc. Third Int. Confi on the Structural Desi~ of Asphalt Pavements. London, 1972. Ann Arbor, 1972 (University of Michigan). 17. NATIONAL INSTITUTE FOR ROAD RESEARCH. The measurement of deflection and curyature of road surfaces. CSIR ManualK16, Pretoria, South Africa 1970. 18. DEHLEN, G L. A simple instrument for measuring the curvature induced in a road surfacing by a 12 wheel load. ~ans., South Africa Inst. of Civil Engineers, Sept, 1962, p. 189. I \\ Elevation I Toe \ Road surface Initial position of wheels in Beam of wheels in Pivot rebound test \ ~- ——— t Effective Ienath of beam Figl DIAGRAMMATIC REPRESENTATION OF THE DEFLECTION BEAM “”.”““I Dimensions a 2.44m b 1.22m c 0.30m d 0.91m 0.61m E;ective length 2.74m “””“ ● 160-180 100-150 ● b 210 (Dimensions in millimetres) Fi~ 2 RANGE OF DIMENSIONS FOR THE CONTACT AREAS OF THE DEFLECTION LORRY TYRES . A Transient deflection ● Transient loading deflection O Transient recovery deflection } Plastic flow of surfacing material L 30 t AGE, 4 MONTHS 20 - A A A A A A 10 L I I 1 I I I 20 25 30 35 40 45 50 30 r I AGE, 4 MONTHS o 0’ 0 0 0 20 - 0 ● 0 ● % ● ‘o x ●, s 10 , 1 1 I 1 I 1 ~ 20 25 30 35 40 45 50 10 I I I I I I 1 20 25 30 35 40 45 50 10 I 1 1 I 1 I I 20 25 30 35 40 45 50 Surfacing temperature at depth of 40mm (OC) . Fig. 3 CHANGE IN TEMPERATURE – DEFLECTION RELATIONSHIP WITH TIME (Overlay 100mm Marshall Asphalt) (UUJ) lle~u!e~ AIqluow o 0 N o 0 4-4 ● 4 0 m o m o 0 ,. ,... Neg. no. B946r19 Mate 1 THE DEFLECTION BEAM AND BEAM FITTED WITH THE SPECIAL SUNSHADES o . “! I —.. Neg. no. B91 9/79 I Plate 2 TOE OF DEFLECTION BEAM IN POSITION ON CALIBRATOR Neg. no. R82180114 Neg. no. R82/80/9 Neg. no.k”82/80/2 Mate3 DEFLECTION BEAM – TEST SEQUENCE , Neg. no. R1 34017718 Plate 4 DEFLECTION BEAM TRUCK FITTED WITH A~USTABLE POINTER I I , Neg. no. R82180118 Plate 5 SHOWING USE OF 2ND BEAM FOR MEASURING FEET MOVEMENT 9. APPENDIX A DEFLECTION SURVEY METHOD FOR EVALUATING ROADS IN DEVELOPING COUNTWES 9.1 Introdudion Deflection surveys are being used increasin~y in developing countries to provide-a basis for overlay thickness design. Ideally a deflection criterion curve will be available for the type of road structure under investigation. From a knowledge of the traffic loadings carried by the pavement in the past and of the deflection and surface condition histories, the remaining ‘life’of the pavement can then be estimated and the most appropriate time for the apphcation of an overlay determined. At the present time it is unlikely that suitable criterion curves exist for most pavements in developing countries. It is also frequently the case that accurate traffic loading histories are unavailable. The use of deflection surveys in developing countries for overlay thickness design ahost invariably impfies that the pavement has already suffered some distress. In these circumstances, without historical data or a criterion curve, the overlay design is based largely upon an assessment of pavement condition, the prediction of the deflection after overlaying, and the ‘life’of the overlay associated with this level of deflection. Pavement condition assessments can be of great value for overlay design purposes, in particular the rate of change of pavement condition is most important. In general a pavement which has a cracked but undeformed surface is basically structurdy sound and only the surfacing has ‘failed’. An untracked but deformed surface indicates that one or more of the lower layers of the pavement has been unable to withstand adequately the apphed traffic stresses, however the rate at which this deformation has developed is important, if it has developed slowly over a number of years it is lessserious than deformation which has occurred rapidy perhaps due to a change in traffic loadings. Analysing the cause of combined deformation and cracking can be difficult. Deformation of the surfacing may have caused it to crack, or cracking of the surfacing, particularly when overlying granular road bases may have a~owed water to enter and weaken the structure causing deformation to occur. If a road is identified as being ~iely to require overlaying in the near future and measurements of deflection and surface condition have not been made previously, then as a minimum requirement, deflection measurements and surface condition assessmentsshould be made well in advance of applying an overlay. Drainage deficiencies should be rectified and badly damaged areas should be patched and made waterproof. Better estimates of reconstruction and overlay thickness requirements can then be made from the results of a second simflar survey carried out after an interval of say one year. During this period axle load surveys should be undertaken. One of the most difficult tasks in planning an overlay programme is the provision of finance. It is frequently the case that overbys are appfied later than the optimal timing because funds are not avaflable at the right time. In addition a length of road which has been included in an overlay programme which is planned to be implemented over a period of two or three years may consequently be excluded from maintenance resealing programmed. Consideration should be given to resealtig pavements even if they are included in an overlay progrme; seals apptied 1 to 2 years before overlaying could wefl produce net savingsin overlay costs or in increased overlay fife. 21 9.2 Defle~ion suwey methods A survey method which has been found suitable for roads showing variable surface conditions has been suggested The method which is reproduced below in a revised form indicates the minimum amount of testing which is required. Deflection measurements should be made in both wheelpaths of the slow lane on dual carriageways and in both lanes of two lane carriageways: 1. 2. 3. 4. Tests on a basic pattern of ten equally spaced tests per kilometre. Additiond tests on any area showing surface distress. If any measurement exceeds a predetermined value, which is related to established deflection criteria for the type of pavement concerned, or indicates the need for a much thicker overlay than is required for the adjacent pavement, the extent of the area involved should be determined by additional tests. Additional tests should be made when the variabihty of the above measurements exceed a certain value. The deflections used both to check variability and to design the overlay should be the largest reading from either wheelpath at each chainage. A simple and adequate check on variabtity can be made in the following way. Ten consecutive measured deflection vahres are considered as a group and their mean value is calculated (for this purpose dl the deflection values at the regular 100m spacings are used but only the ma~mum reading is taken from any one area tested on account of its surface distress or large deflection). If the largest and smallest values of the group differ from the mean by more than one-third of the mean then four addhional tests should be made, each test being made midway between the test points which registered the largest and smdest deflections and their immediate adjacent test points. If any of these four measurements would reduce the distante between test points to lessthan 5m then the extra measurement may be omitted. The process is then repeated on another group of ten consecutive deflection values which is formed by taking into the previous group the next deflection value along the road, whflst omitting the first deflection value from the previous group. The process is repeated as each new deflection measurement is made so that a running check on variability is maintained. The aim should be,to have at least 20 measurements per kilometre spaced in such a way that they reflect the variation in pavement conditions. At least 20 measurements per kilometre should be made on pavements which do not show any surface distress. If time is available measurements should be taken at closer intervals which may increase the accuracy of the method. Testing at 12m intervals is recommended for surveys in the United figdorn7. After dl measurements have been made it is convenient to plot for each lane the maximum deflection (corrected for temperature effect) and the worst pavement surface condition ratings for each cross-section against chainage along the road site. Any area showing exceptional weakness which may require reconstruction or specialtreatment is delineated. The remainder of the road is then divided into sections by inspection of 22 the plot in such a way as to minimise variation in deflections withti each section. a section should be compatible with the method of resurfacing that it is intended The minimum length of to employ, bearing in mind the frequency of thickness adjustments which can sensibly be made to the paving machine. If deflection values are very high, implying that pavement reconstruction is required, then short lengths of pavement can be considered separately since normafly hand labour (assisted by machine) would be used for such an operation. The find stage is to design the overhy for each section of the road and to determine those areas which merit further investigation. On those areas of pavement where deflections are very high or fadure hs occurred inspection holes should be dug and a careful note of drtiage conditions made. If it is assumed that the distribution of deflections is normal then only 2.5 per cent of the pavement would be expected to have a deflection in excess of the sample mean plus twice the standard deviation of the sample. The survey method proposed wi~ tend to separate out the very high deflections on areas warranting special treatment or reconstruction and so reduce the effect of skewness,thus the distribution of deflections in each selected section should not be far removed from a normal distribution. , Some authorities recommend an overlay design deflection of the mean plus twice the standard deviation of a sample of readings. It is felt that in developing countries it is not appropriate to go to the expense of such a high degree of certainty which might lead to the expensive over-design of large areas of pavement. Use of the proposed survey method in conjunction with a design deflection of the mean plus 1.0 to 1.5 standard deviations for the section is Recommended. 23 . (885) Dd0536380 1;500 5/80 HPLtd So’ton G1915 PRINTED IN ENGLAND ABSTRACT Measurement of pavement deflections in tropical and sub-tropical climates: H R SMITH and C R JONES: Department of the Environment Department of Transport, TRRL Laboratory Report 935: Crowthorne, 1980 (Transport and Road Research Laboratory). The measurement of the deflection of the surface of road pavements under a standard wheel load is common practice in many countries. The main purposes of making deflection measurements are to assist in the evaluation of the structural condition of roads, and to help in the design of road strengthening measures. In developing countries in the tropics the high road temperatures and other climatic effects necessitate the use of a special technique in making deflection tests. This report describes the. technique adopted by the Overseas Unit of the Transport and Road Research Laboratory for the measurement of pavement deflection in tropical developing countries, and discusses factors relevant to the deflection of road pavements in such conditions. ISSN 0305– 1293 ABSTRACT Measurement of pavement deflections in tropical and sub-tropical climates: H R SMITH and C R JONES: Department of the Environment Department of Transport, TRRL Laboratory Report 935: Crowthorne, 1980 (Transport and Road Research Laboratory). The measurement of the deflection of the surface of road pavements under a standard wheel load is common practice in many countries. The main purposes of making deflection measurements are to assist in the evaluation of the structural condition of roads, and to help in the design o.froad strengthening measures. In developing countries in the tropics the high road temperatures and other climatic effects necessitate the use of a special technique in making deflection tests. This report describes the technique adopted by the Overseas Unit of the Transport and Road Research Laboratory for the measurement of pavement deflection in tropical developing countries, and discusses factors relevant to the deflection of road pavements in such conditions. ISSN 0305–1293