Home

Roughness Manual - Roads and Highways Department

1. Roughness Surface scale type 1 A 2 IRI All 0 593 m km 0 185 BI All 983 mm km 423 BI AC 574 mm km 401 BI ST 132 mm km 220 BI GR 1134 mm km 676 BI EA 2230 mm km 797 1 AC Asphaltic concrete ST a Surface treated GR Gravel EA Earth 2 Bracketed values are one standard error 200 observations was chosen as a practical upper limit from the point of view of managing the data handling and limiting the length of time taken to measure D A 3 1 Choice of machine length The choice of machine length was examined by simulat ing Merlins of lengths ranging from 0 6 to 3 metres Using the same procedure as that described above and not distinguishing between the different types of surface linear regressions were derived relating the value of roughness on the two measuring scales to D for each Merlin length Figure A5 shows the R values for these regressions On the IRI scale the best correlations are between 1 4 and 2 6 metres The highest value occurs at around 1 8 metres and so this was chosen as the standard Merlin length Reducing the length below 1 4 metres causes a sharp decrease in correlation Turning to the results for the BI scale the answer is quite different Here the best correlation is more sharply defined and occurs at a Merlin length of one metre The degree of correlation is not as good as the best IRI value but
2. 9 5 1 2 Survey Methodology eri beoe 10 6 DATA ENTRBY icc dcc iu ruat ee 11 Picture 1 Bump Integraltor eor pet ec rra iae pedet ep 4 Picture 2 Counter Unit Bump Integrator 4 Picture Distance 5 Picture 4 Merlin in 5 Picture 5 Merlin Worksheet eeseeeeeee ennemi nan 8 Picture 6 Regression analysis chart eee 9 Appendix 1 CNS Farnell Distance Odometer Appendix 2 Merlin Manual E Documents Roughness Manual oad Roughness Survey Manual docPage 2 1 1 Ministry of Communication Road Roughness Roads and Highways Department Survey Manual 1 INTRODUCTION Maintenance of road is dependent on several factors one of which is the condition of the pavement surface To determine what treatment is necessary the condition of the surface is to be known Roughness is the measurement of riding quality which in turn is the effect of total surface deterioration In Roads and Highways Department RHD the HDM Circle has the responsibility to produce the priority list of the Annual Road Maintenance Programme using HDM 4 software Roughness of roads in terms of IRI has the greatest influence in the HDM analysis So correct measurement of Roughness is a prerequisite for doing the correct HDM
3. it will be found that when the machine is tilted from side to side the pointer moves When correctly adjusted leaning the machine over to one side so that the stabiliser rests on the road has little effect on the position of the pointer Before use the mechanical amplification of the arm should be checked using a small calibration block typically 6 mm thick Insertion of the block under the probe should move the pointer by 60 mm and any discrepancy has to be allowed for For example if the pointer moved by only 57 mm then the value of D measured on the chart should be increased by a factor of 60 57 It is also recommended that a check is carried out before and after each set of measurements to ensure that there has been no unwanted movement of critical parts such as the rear foot or the probe mounting The check is carried out by returning the machine to a precisely defined position along the road and making sure that the same pointer reading is obtained If when making measurements on a very rough road more than 10 readings are at either limit of the histogram the probe should be removed and attached to the alternative fixing point which is provided This is twice as far from the pivot and reduces the mechanical amplifica tion of the arm to 5 halving the width of the distribution Values of D read from the chart are scaled using the calibration procedure described earlier Although the spacing between the probe and the two feet
4. 270 mm km on the BI scale for the best asphaltic concrete surface to 15 91 m km 16 750 mm km on the scale for the worst earth surface Figure A1 shows as an example the road profile as measured by the Abay Beam along 50 metres of two of the test sections The first is an asphaltic concrete road in relatively good condition while the second is a gravel surface in fair condition As might be expected compared to the asphaltic concrete the gravel surface shows a much greater presence of short wavelength undulations To help visualise the Merlin s response the Figure also shows the machine s length 1 8 metres on the same scale A 2 SIMULATION RESULTS Given these road profiles it was possible to carry out a computer simulation of the performance of a Merlin Neglecting the small effects due to the fact that the Merlin is not operated in a horizontal position if it is assumed that the rear foot is placed at a horizontal distance of X metres from the start of the section then the probe would be at a distance of X 0 9 metres from the start and the front foot at a distance of X 1 8 metres If the corresponding vertical distances at these points are Yo Y and then the pointer on the Merlin will be displaced from the zero position by an amount d given by M x Y 0 5 x Y Y 1 where M is the mechanical amplification provided by the moving arm usually close to 10 Placing the Merlin at successive posi
5. If working to the BI scale and using a single relationship for all surface types systematic errors are much larger The RMS residual error for single measure ments was 21 per cent and this reduced only slightly to 19 per cent for four measurements The benefits of multiple measurements are greater when using separate relationships for each surface type the RMS residual errors ranged from 9 to 15 per cent for single measurements compared with 4 to 13 per cent for multiple measurements The relatively large error for asphaltic concrete compared to surface treated roads could well reflect the more limited roughness range for the latter and that the true relationships are non linear When estimating roughness for a vehicle the normal procedure is to assume that the combined roughness for the two wheel tracks can be equated to the mean of the individual tracks although this does give rise to a small error Hence in practice a minimum of two sets of Merlin observations are required The roughness of the individ ual wheel tracks can differ considerably Bearing in mind the above limitations it is normally better to calibrate an RTRRMS device at a larger number of sites than make many repeat measurements at the same site Moreover particularly if working on the BI scale these sites should have similar surfaces to those on which the RTRRMS is to be used A number of other practical points should be considered when measuring roughness or
6. analysis In addition the HDM Circle of RHD is responsible to do the roughness measurement of all paved roads within the RHD Road Network This instruction manual has been prepared solely for use within RHD and describes the measurement methods that are being adopted for the road roughness measurement This guide is accessible on the RHD Intranet 2 TIMING OF ROUGHNESS MEASUREMENT Roughness measurement is done annually preferably from October to December To measure the roughness of all paved roads takes about three months if 3 nos of vehicles are used for the survey 3 EQUIPMENT NEEDED Following equipment are needed for the roughness measurement Vehicles Pickup or Jeep Bump Integrator Distance Measurement Equipment CSN Farnell TRL MERLIN Other materials and stationers required are Paint and brush Graph papers cm scale Wooden pencils Jute ropes 3 4 DESCRIPTION OF EQUIPMENT 3 1 1 Vehicles Generally pickup and Jeep are well suited for the survey Bump integrator is fitted on the vehicle just above the rear axle A steel hook is welded to the differential so that a E Documents Roughness Manual oad Roughness Survey Manual docPage 3 1 1 Ministry of Communication Road Roughness Roads and Highways Department Survey Manual 1 INTRODUCTION Maintenance of road is dependent on several factors one of which is the condition of the pavement surface To determine what treatment is necessary the condition o
7. and this gives rise to the systematic differences mentioned above The relationships between the Merlin and the IRI scales are very similar for all the surface types examined whereas the relationships between the Merlin and the BI scales and the IRI and scales are clearly different for each surface type This implies that the effective spectral sensitivity of the Merlin is closer to that of the IRI scale than the BI scale It is interesting to note that the coeffi cients and constants in equations 3 to 6 follow a steady progression as the surfaces vary from asphaltic concrete to earth presumably reflecting a progressive change in spectral signature When the random error is greater than the systematic error significant improve ments can be made by repeat measurements on the same road section If the systematic error increases the benefit of repeat measurements on the same section decreases Table 1 which was derived from the com puter simulation shows the mean residual error in roughness level for estimates based on one and four runs of the Merlin If roughness is being measured directly on the Merlin scale then there are no systematic errors to contend with and the error falls with the reciprocal of the square root of the number of observations As Table 1 shows a single measurement gave a root mean square RMS residual error of 8 per cent while taking the mean of four observa tions halved the error to 4 per cent If
8. calibrating an RTRRMS and a useful guide is provided by Sayers et al 1986b As a simple cross check on performance roughness values on the Merlin scale were measured for a series of asphaltic concrete test sections on the TRRL experimen tal track Four measurements were taken on each section and the mean values are shown plotted in Figure 6 against the roughness of each section on the BI scale as measured with the Abay beam Abaynayaka 1984 The Figure also shows the Merlin BI calibration line for asphaltic concrete roads as given in equation 3 As can be seen the points lie very close to the calibration line and while the check is by no means comprehensive it does lend strong support to the results derived from the simulation 6 DISCUSSION The reason for designing the Merlin was to provide a device which is easy to use and reasonably accurate and yet can be manufactured and maintained with the limited resources available within developing countries Experi ence indicates that it has been successful in meeting these objectives A number of the machines have been made at TRRL and shipped overseas while other units have been made overseas from drawings provided by the Laboratory To date Merlins have been used in 11 developing countries in South America Africa and Asia in six of these the equipment was made locally at current prices of typically 250 US dollars One inconvenience of the Merlin is that because of its length it
9. is not easily transported within a vehicle A shorter machine could be used but as is shown in the Appendix this will lead to some reduction in correlation with the IRI scale Alternatively a more portable design could be considered using a structure which folds or dismantles While this is a possibility it has been avoided because of the need to retain rigidity Although its design is very simple the Merlin is able to measure displace ments to less than a millimetre and this ability could easily be compromised by unwanted flexing of the structure In recent years there has been a move towards reducing the number of different roughness scales in use and standardising on the International Roughness Index However the Merlin scale does have the advantage of being easy to visualise and although Merlin readings can be converted easily to IRI values in some cases this conversion is unnecessary and direct use of the Merlin scale should be considered 7 ACKNOWLEDGEMENTS This work forms part of the programme of research of the Overseas Unit Head J S Yerrell of the Transport and Road Research Laboratory UK 8 REFERENCES ABAYNAYAKA S W 1984 Calibrating and standardis ing road roughness measurements made with response type instruments In ENPC International Conference on Roads and Development Paris May 1984 pp13 18 Presses de l ecole nationale des ponts et chaussees Paris CHESHER A and HARRISON R 1987 Vehicle
10. measuring roughness on the IRI scale taking four measurements gave an RMS residual error of 7 per cent compared with 10 per cent when using single measure TABLE 1 Residual errors _ OO eee One observation RMS residual error Four observations Roughness Surface scale type Merlin All mm IRI All m km BI All mm km BI AC mm km BI ST mm km BI GR mm km BI EA mm km AC Asphaltic concrete ST Surface treated GR Gravel EA Earth 8 4 10 7 21 19 15 13 9 4 12 11 ments If working to the BI scale and using a single relationship for all surface types systematic errors are much larger The RMS residual error for single measure ments was 21 per cent and this reduced only slightly to 19 per cent for four measurements The benefits of multiple measurements are greater when using separate relationships for each surface type the RMS residual errors ranged from 9 to 15 per cent for single measurements compared with 4 to 13 per cent for multiple measurements The relatively large error for asphaltic concrete compared to surface treated roads could well reflect the more limited roughness range for the latter and that the true relationships are non linear When estimating roughness for a vehicle the normal procedure is to assume that the combined roughness for the two wheel tracks can be equated to the mean of t
11. of measurement Discussion Acknowledgements p mom s References Appendix A Simulation of performance A 1 The International Road Roughness Experiment A 2 Simulation results A 3 Alternative procedures and designs A 3 1 Choice of machine length A 3 2 Measurement of data spread Page gt O N N 10 10 10 12 12 12 15 17 17 THE MERLIN LOW COST ROAD ROUGHNESS MEASURING MACHINE ABSTRACT The roughness of a road s surface is an important measure of road condition and a key factor in determining vehicle operating costs on poor quality surfaces This report describes a simple roughness measuring machine which has been designed especially for use in developing countries It is called MERLIN a Machine for Evaluating Roughness using Low cost INstrumentation The device can be used either for direct measurement or for calibrat ing response type instruments such as the vehicle mounted bump integrator It consists of a metal frame 1 8 metres long with a wheel at the front a foot at the rear and a probe mid way between them which rests on the road surface The probe is attached to a moving arm at the other end of which is a pointer which moves over a chart The machine is wheeled along the road and at regular intervals the position of the pointer is recorded on the chart to build up a histogram The width of this histogram can be used to give a good estimate of roughness in terms of the International Rou
12. operating costs evidence from developing countries John Hopkins University Press Baltimore and London CRRI 1982 Road user cost study in India final report Central Road Research Institute New Delhi FACE COMPANY The Edward W Face Company Inc Norfolk Virginia GILLESPIE T D 1986 Developments in road rough ness measurement and calibration procedures In ARRB Proc 13th ARRB 5th REAAA Conf 13 1 pp 91 112 Australian Road Research Board Vermont South HIDE H 1982 Vehicle operating costs in the Carib bean results of a survey of vehicle operators TRRL Laboratory Report 1031 Transport and Road Research Laboratory Crowthorne HIDE H et al 1975 The Kenya road transport cost study research on vehicle operating costs TRRL Laboratory Report 672 Transport and Road Research Laboratory Crowthorne ments If working to the BI scale and using a single relationship for all surface types systematic errors are much larger The RMS residual error for single measure ments was 21 per cent and this reduced only slightly to 19 per cent for four measurements The benefits of multiple measurements are greater when using separate relationships for each surface type the RMS residual errors ranged from 9 to 15 per cent for single measurements compared with 4 to 13 per cent for multiple measurements The relatively large error for asphaltic concrete compared to surface treated roads could well reflect the more li
13. or for calibrat ing response type instruments such as the vehicle mounted bump integrator It consists of a metal frame 1 8 metres long with a wheel at the front a foot at the rear and a probe mid way between them which rests on the road surface The probe is attached to a moving arm at the other end of which is a pointer which moves over a chart The machine is wheeled along the road and at regular intervals the position of the pointer is recorded on the chart to build up a histogram The width of this histogram can be used to give a good estimate of roughness in terms of the International Roughness Index Calibration of the device was carried out using computer simulations of its operation on road profiles measured in the 1982 International Road Roughness Experiment Merlins are in use in a number of developing countries They can usually be made locally at a current cost of typically 250 US 1 INTRODUCTION The longitudinal unevenness of a road s surface nor mally termed its roughness is both a good measure of the road s condition and an important determinant of vehicle operating costs and ride quality Within develop ing countries there is particular interest in the effect on vehicle operating costs A number of studies Hide et al 1975 Hide 1982 CRRI 1982 Chesher amp Harrison 1987 have shown how roughness can influence the cost of vehicle maintenance the extent of tyre damage and vehicle running speeds and hence vehicle
14. rim of the wheel and all measurements are taken with the mark at its closest proximity to the road The wheel is then said to be in its normal position 3 3 METHOD OF USE The recommended procedure to determine the rough ness of a stretch of road is to take 200 measurements at regular intervals say once every wheel revolution At each measuring point the machine is rested on the road with the wheel in its normal position and the rear foot probe and stabiliser in contact with the road surface The operator then records the position of the pointer on the chart with a cross in the appropriate column and to keep a record of the total number of observations makes a cross in the tally box on the chart Moving arm Stablliser Figure 2 Sketch of the Merlin The handles of the Merlin are then raised so that only the wheel remains in contact with the road and the machine is moved forward to the next measuring point where the process is repeated The spacing between the measuring points does not matter as long as the readings are always taken with the wheel in the normal position Taking measurements at regular intervals should both produce a good average sample over the whole length of the section and reduce the risk of bias due to the opera tor tending to avoid particularly bad sections of road Figure 3 shows a typical completed chart When the 200 observations have been made the chart is removed from the Merlin The positions
15. roughness measuring machines and to calibrate their measures to a common scale As part of this study the machines were run over a series of test sections 320 metres long for four types of road surface asphaltic concrete surface treated gravel and earth One of the instruments used in the Study was an early version of the TRRL Abay Beam This employed an aluminium beam 3 metres in length supported at each end by adjustable tripods which were used for levelling Running along the beam was a sliding carriage which had at its lower end a wheel of 250 mm diameter which was in contact with the road surface A linear transducer inside the carriage measured the distance between the bottom of the wheel and the beam to the nearest milli metre and this was recorded at 100 mm intervals along the road By successively relocating the beam along the length of the road section and repeatedly levelling the beam the recordings provided a continuous sampling of the road profile Data from the Abay beam were available for 27 of the test wheel paths These are listed in Table A1 together with roughness on the IRI scale as computed from the beam road profile data and roughness on the BI scale as measured by a fifth wheel bump integrator towed at 32 km h As can be seen there are eight paths on asphaltic concrete roads five on surface treated roads seven on gravel surfaces and seven on earth surfaces Rough nesses range from 2 44 m km on the IRI scale 1
16. 1031 Transport and Road Research Laboratory Crowthorne HIDE H et al 1975 The Kenya road transport cost study research on vehicle operating costs TRRL Laboratory Report 672 Transport and Road Research Laboratory Crowthorne 4 000 5 000 t E 2 000 E co 1 000 Calibration relationship 574 29 90 O 20 40 60 80 100 Merlin D mm Figure 6 Calibration check on asphaltic concrete JORDAN G and YOUNG J C 1980 Developments in the calibration and use of the Bump Integrator for ride assessment TRRL Supplementary Report 604 Trans port and Road Research Laboratory Crowthorne SAYERS W S et al 1986a The International Road Roughness Experiment establishing correlation and a calibration standard for measurements World Bank Technical Paper Number 45 The World Bank Washing ton D C SAYERS W S et al 1986b Guidelines for conducting and calibrating road roughness measurements World Bank Technical Paper Number 46 The World Bank Washington D C STILL P B and JORDAN P G 1980 Evaluation of the TRRL high speed profilometer TRRL Laboratory Report 922 Transport and Road Research Laboratory Crowthorne APPENDIX A SIMULATION OF PERFORMANCE A 1 THE INTERNATIONAL ROAD ROUGHNESS EXPERIMENT In 1982 a major study the International Road Rough ness Experiment IRRE was carried out in Brasilia Sayers et al 1986a to compare the performance of a number of different road
17. Government of the People s Republic of Bangladesh Ministry of Communications Roads and Highways Department ROAD ROUGHNESS SURVEY MANUAL OCTOBER 2001 Ministry of Communication Road Roughness Roads and Highways Department Survey Manual TABLE OF CONTENTS A INTRODUGTION 2 3 2 TIMING OF ROUGHNESS MEASUREMENT erre enne nnn nna 3 3 EQUIPMENT NEEDED 3 3 1 DESCRIPTION OF EQUIPMENT eere 3 Behl Vehicles os a tossed ECL me EIR E aer it iU HALLA mE 3 3 1 2 BumpJIntegralor sco pretatur ERR el arte 4 3 1 3 Distance Measurement Equipment esee 5 VUE ERU 5 4 ACTIVITIES INVOLVED e eere nnne nnn nn nn nana nuns r iara dana ara aa RR RR sona 6 4 1 CALIBRATION OF CNS FARNELL DISTANCE ODOMETER 6 4 2 DOING THE MERLIN EXERCISE ccccccecccceeseceeeeeeeceeeeseeeeaueeeeceeeaseueeaueeaeceeeaneuees 6 4 2 1 Selection of sile a a a a a a laa i 7 4 2 2 Doing the MERLIN SUIVOV ac o apio e etuer ee repente 7 42 3 Running Vehicles on Merlin Sites esses 7 4 2 4 Doing the Regression Analysis sss eise 8 5 DOING THE ACTUAL ROUGHNESS SURVEY 9 5At TOTIS m ced ede dU ad deba NIA Bode aden one mie drei nta
18. I values in some cases this conversion is unnecessary and direct use of the Merlin scale should be considered 7 ACKNOWLEDGEMENTS This work forms part of the programme of research of the Overseas Unit Head J S Yerrell of the Transport and Road Research Laboratory UK 8 REFERENCES ABAYNAYAKA S W 1984 Calibrating and standardis ing road roughness measurements made with response type instruments In ENPC International Conference on Roads and Development Paris May 1984 pp13 18 Presses de l ecole nationale des ponts et chaussees Paris CHESHER A and HARRISON R 1987 Vehicle operating costs evidence from developing countries John Hopkins University Press Baltimore and London CRRI 1982 Road user cost study in India final report Central Road Research Institute New Delhi FACE COMPANY The Edward W Face Company Inc Norfolk Virginia GILLESPIE T D 1986 Developments in road rough ness measurement and calibration procedures In ARRB Proc 13th ARRB 5th REAAA Conf 13 1 pp 91 112 Australian Road Research Board Vermont South HIDE H 1982 Vehicle operating costs in the Carib bean results of a survey of vehicle operators TRRL Laboratory Report 1031 Transport and Road Research Laboratory Crowthorne HIDE H et al 1975 The Kenya road transport cost study research on vehicle operating costs TRRL Laboratory Report 672 Transport and Road Research Laboratory Crowthorne ments
19. Merlin measurement As an example if the ratio of the sections of the moving arm is 9 75 then the movement of the pointer will be less by a factor of 0 975 9 75 10 Therefore the width of the histogram D will have to be increased by a multiplying factor of 1 03 10 9 75 to get the correct roughness Picture 4 Merlin in use E Documents Roughness Manual Road Roughness Survey Manual docPage 5 1 1 Ministry of Communication Road Roughness Roads and Highways Department Survey Manual 4 ACTIVITIES INVOLVED The roughness survey requires several activities of which each one has the effect on the other The activities are as follows 4 1 Calibration of CNS Farnell Distance Odometer During the actual survey bump readings are noted after travelling every 0 5 km of road Therefore a correct calibration of the distance odometer is required See the enclosed CNS Farnell Distance Odometer Manual for instructions 4 2 Doing the Merlin Exercise This is done to get the correct roughness of a particular portion of a road so that if a vehicle fitted with a Bump Integrator is run over that portion the relative bump counts for the known roughness could be known In this way if the relative roughness and bump counts of say 10 sites of different roughness are known then a relationship between the roughness and bumps can be established by doing a regression analysis In RHD generally 10 to 11 sites are surveyed with Merlin to obtain the
20. PERATION The principle of operation is as follows The device has two feet and a probe which rest on the road surface along the wheel track whose roughness is to be measured The feet are 1 8 metres apart and the probe lies mid way between them see Figure 1 The device measures the vertical displacement between the road surface under the probe and the centre point of an imaginary line joining the two points where the road surface is in contact with the two feet This displacement is known as the mid chord deviation If measurements are taken at successive intervals along road then the rougher the road surface the greater the variability of the displacements By plotting the displace ments as a histogram on a chart mounted on the instru ment it is possible to measure their spread and this has been found to correlate well with road roughness as measured on standard roughness scales The concept of using the spread of mid chord deviations as a means of assessing road roughness is not new For example two roughness indices Ql and MO have been proposed by other researchers and are described by Sayers et al 1986a They are each based on the root Figure 1 Measurement of mid chord deviation mean square values of two mid chord deviations with different base lengths and have been suggested as standards which can be calculated relatively easily from road profiles measured by rod and level However the Merlin operates by u
21. actual roughness of individual site The sites are initially selected in such a way to give varying roughness low and high Each survey vehicle is then run on those sites to obtain the bump counts against the corresponding roughness of sites and lastly the regression analysis is done to get the relationship of roughness and bump counts of individual vehicle E Documents Roughness Manual Road Roughness Survey Manual docPage 6 1 1 Ministry of Communication Road Roughness Roads and Highways Department Survey Manual 4 2 1 Selection of site Selection of site is very important because sites should be of varying roughness low and high It may be difficult to get the road with low roughness and similar the case with high roughness However from the experience it was seen that roads of low roughness like 2 00 IRI were available and on the high range 8 00 to 8 50 or 9 00 were also available Use the bump integrator and the distance odometer during site selection When driving upon a test site have the bump integrator on and keep the distance odometer open Note down the bump counts for every 0 5 km The more the bump counts the more will be the possible roughness 4 2 2 Doing the MERLIN Survey Generally two Merlins are used for the survey Put a mark on the rod with paint at the starting Place the Merlins side by side so that each Merlin is on the wheel track the outer and the inner Always follow the left side of the road One vehicle shoul
22. ator 2 ROUGHNESS MEASURING INSTRUMENTS Roughness measuring instruments can be grouped into three different classes The simplest in concept are the static road profile measuring devices such as the rod and level which measure surface undulations at regular intervals Unfortunately these devices are very slow in use and there can be a considerable amount of calcula tion involved in deriving roughness levels from the measurements taken Two recent devices which work on a similar principle but are semi automated are the TRRL Abay beam Abaynay aka 1984 and the modified Dipstick profiler Face Company With both of these instruments the surface undulations are measured from a static reference and data is fed directly into a microprocessor to do the necessary calculations They produce high quality results but they are relatively slow in operation and expensive The second class of instrument is the dynamic profile measuring device such as the TRRL high speed profil ometer Still and Jordan 1980 In these instruments surface undulations are measured with respect to a moving platform equipped with some means of compen sating for platform movement so that the true road profile can be derived This is then converted to roughness indices by automatic data processing These devices can operate at high speeds and give good quality results but they are very expensive they are not usually suitable for very rough roads and they ha
23. bly accurate and yet can be manufactured and maintained with the limited resources available within developing countries Experi ence indicates that it has been successful in meeting these objectives A number of the machines have been made at TRRL and shipped overseas while other units have been made overseas from drawings provided by the Laboratory To date Merlins have been used in 11 developing countries in South America Africa and Asia in six of these the equipment was made locally at current prices of typically 250 US dollars One inconvenience of the Merlin is that because of its length it is not easily transported within a vehicle A shorter machine could be used but as is shown in the Appendix this will lead to some reduction in correlation with the IRI scale Alternatively a more portable design could be considered using a structure which folds or dismantles While this is a possibility it has been avoided because of the need to retain rigidity Although its design is very simple the Merlin is able to measure displace ments to less than a millimetre and this ability could easily be compromised by unwanted flexing of the structure In recent years there has been a move towards reducing the number of different roughness scales in use and standardising on the International Roughness Index However the Merlin scale does have the advantage of being easy to visualise and although Merlin readings can be converted easily to IR
24. d be at the back of the survey team and the other at the front with flags for the safety reason Fix graph sheets on the steel plates of the Merlins over which the pointer moves Now move the Merlin forward one complete rotation of the wheel see the position of the pointer and put a cross mark on the graph sheet In this way after every rotation of the Merlin put a mark on the graph sheet at the position of the pointer Continue doing this for 200 rotations of the Merlin wheel and again put a mark on the road by drawing a line by paint To keep a correct counting of the no of rotations a tally box of 200 small squares of the graph sheet can be utilized After every rotation one square can be marked so that after 200 rotations the 200 squares of the tally box will be completely marked When 200 rotations will be completed a histogram like the one shown below will be built in the graph paper The positions mid way between the tenth and the eleventh crosses counting in from each end of the distribution are marked on the chart below the columns It may be necessary to interpolate between column boundaries as shown in the figure below The spacing between the two marks D is then measured in millimeters and this is the roughness on the Merlin scale It is to be noted that the final D value will be the average of D values obtained from two Merlins Once the final D value is obtained substitute it the standard formula given the Merlin guide to get the roughne
25. d use of the Bump Integrator for ride assessment TRRL Supplementary Report 604 Trans port and Road Research Laboratory Crowthorne SAYERS W S et al 1986a The International Road Roughness Experiment establishing correlation and a calibration standard for measurements World Bank Technical Paper Number 45 The World Bank Washing ton D C SAYERS W S et al 1986b Guidelines for conducting and calibrating road roughness measurements World Bank Technical Paper Number 46 The World Bank Washington D C STILL P B and JORDAN P G 1980 Evaluation of the TRRL high speed profilometer TRRL Laboratory Report 922 Transport and Road Research Laboratory Crowthorne APPENDIX A SIMULATION OF PERFORMANCE A 1 THE INTERNATIONAL ROAD ROUGHNESS EXPERIMENT In 1982 a major study the International Road Rough ness Experiment IRRE was carried out in Brasilia Sayers et al 1986a to compare the performance of a number of different road roughness measuring machines and to calibrate their measures to a common scale As part of this study the machines were run over a series of test sections 320 metres long for four types of road surface asphaltic concrete surface treated gravel and earth One of the instruments used in the Study was an early version of the TRRL Abay Beam This employed an aluminium beam 3 metres in length supported at each end by adjustable tripods which were used for levelling Running along the beam was a sli
26. ding carriage which had at its lower end a wheel of 250 mm diameter which was in contact with the road surface A linear transducer inside the carriage measured the distance between the bottom of the wheel and the beam to the nearest milli metre and this was recorded at 100 mm intervals along the road By successively relocating the beam along the length of the road section and repeatedly levelling the beam the recordings provided a continuous sampling of the road profile Data from the Abay beam were available for 27 of the test wheel paths These are listed in Table A1 together with roughness on the IRI scale as computed from the beam road profile data and roughness on the BI scale as measured by a fifth wheel bump integrator towed at 32 km h As can be seen there are eight paths on asphaltic concrete roads five on surface treated roads seven on gravel surfaces and seven on earth surfaces Rough nesses range from 2 44 m km on the IRI scale 1 270 mm km on the BI scale for the best asphaltic concrete surface to 15 91 m km 16 750 mm km on the scale for the worst earth surface Figure A1 shows as an example the road profile as measured by the Abay Beam along 50 metres of two of the test sections The first is an asphaltic concrete road in relatively good condition while the second is a gravel surface in fair condition As might be expected compared to the asphaltic concrete the gravel surface shows a much greater presence of
27. ehicle operating costs and pavement deterioration is the output of the fifth wheel BI towed at 32 km h How ever another scale which is now being widely used is the International Roughness Index Sayers et al 1986a This scale which is derived from road profile data by a fairly complex mathematical procedure represents the vertical movement of a wheel with respect to a chassis in an idealised suspension system when travelling along the road at 80 km h As with the BI scale it is measured in terms of units of vertical movement of the wheel per unit length of road and is normally quoted in metres per kilometre Traditionally the BI scale is normally quoted in millimetres per kilometre 3 THE MERLIN The new instrument which has been developed is a variation of the static profile measuring device It is a manually operated instrument which is wheeled along the road and measures surface undulations at regular intervals Readings are easily taken and there is a graphical procedure for data analysis so that road roughness can be measured on a standard roughness scale without the need for complex calculation Its particular attractions for use in the developing world are that it is robust inexpensive simple to operate and easy to make and maintain The device is called MERLIN which is an acronym for a Machine for Evaluating Roughness using Low cost INstrumentation It was designed on the basis of a computer simulation of its operati
28. f the surface is to be known Roughness is the measurement of riding quality which in turn is the effect of total surface deterioration In Roads and Highways Department RHD the HDM Circle has the responsibility to produce the priority list of the Annual Road Maintenance Programme using HDM 4 software Roughness of roads in terms of IRI has the greatest influence in the HDM analysis So correct measurement of Roughness is a prerequisite for doing the correct HDM analysis In addition the HDM Circle of RHD is responsible to do the roughness measurement of all paved roads within the RHD Road Network This instruction manual has been prepared solely for use within RHD and describes the measurement methods that are being adopted for the road roughness measurement This guide is accessible on the RHD Intranet 2 TIMING OF ROUGHNESS MEASUREMENT Roughness measurement is done annually preferably from October to December To measure the roughness of all paved roads takes about three months if 3 nos of vehicles are used for the survey 3 EQUIPMENT NEEDED Following equipment are needed for the roughness measurement Vehicles Pickup or Jeep Bump Integrator Distance Measurement Equipment CSN Farnell TRL MERLIN Other materials and stationers required are Paint and brush Graph papers cm scale Wooden pencils Jute ropes 3 4 DESCRIPTION OF EQUIPMENT 3 1 1 Vehicles Generally pickup and Jeep are well suited for the survey Bu
29. ghness Index Calibration of the device was carried out using computer simulations of its operation on road profiles measured in the 1982 International Road Roughness Experiment Merlins are in use in a number of developing countries They can usually be made locally at a current cost of typically 250 US 1 INTRODUCTION The longitudinal unevenness of a road s surface nor mally termed its roughness is both a good measure of the road s condition and an important determinant of vehicle operating costs and ride quality Within develop ing countries there is particular interest in the effect on vehicle operating costs A number of studies Hide et al 1975 Hide 1982 CRRI 1982 Chesher amp Harrison 1987 have shown how roughness can influence the cost of vehicle maintenance the extent of tyre damage and vehicle running speeds and hence vehicle productivity Reliable measurement of road roughness is therefore seen as an important activity in road network manage ment Several different road roughness scales have been established and a variety of roughness measuring machines have been developed However it was felt that there was a need particularly within developing coun tries for a new simple type of measuring instrument which could be used either directly to measure roughness over a limited part of the road network or for calibrating other roughness measuring equipment particularly the very widely used vehicle mounted bump integr
30. he individual tracks although this does give rise to a small error Hence in practice a minimum of two sets of Merlin observations are required The roughness of the individ ual wheel tracks can differ considerably Bearing in mind the above limitations it is normally better to calibrate an RTRRMS device at a larger number of sites than make many repeat measurements at the same site Moreover particularly if working on the BI scale these sites should have similar surfaces to those on which the RTRRMS is to be used A number of other practical points should be considered when measuring roughness or calibrating an RTRRMS and a useful guide is provided by Sayers et al 1986b As a simple cross check on performance roughness values on the Merlin scale were measured for a series of asphaltic concrete test sections on the TRRL experimen tal track Four measurements were taken on each section and the mean values are shown plotted in Figure 6 against the roughness of each section on the BI scale as measured with the Abay beam Abaynayaka 1984 The Figure also shows the Merlin BI calibration line for asphaltic concrete roads as given in equation 3 As can be seen the points lie very close to the calibration line and while the check is by no means comprehensive it does lend strong support to the results derived from the simulation 6 DISCUSSION The reason for designing the Merlin was to provide a device which is easy to use and reasona
31. ill not be central The lateral position of the probe has to be adjusted so that its traverse passes centrally through the line joining the bottom of the tyre and the rear foot If not it will be found that when the machine is tilted from side to side the pointer moves When correctly adjusted leaning the machine over to one side so that the stabiliser rests on the road has little effect on the position of the pointer Before use the mechanical amplification of the arm should be checked using a small calibration block typically 6 mm thick Insertion of the block under the probe should move the pointer by 60 mm and any discrepancy has to be allowed for For example if the pointer moved by only 57 mm then the value of D measured on the chart should be increased by a factor of 60 57 It is also recommended that a check is carried out before and after each set of measurements to ensure that there has been no unwanted movement of critical parts such as the rear foot or the probe mounting The check is carried out by returning the machine to a precisely defined position along the road and making sure that the same pointer reading is obtained If when making measurements on a very rough road more than 10 readings are at either limit of the histogram the probe should be removed and attached to the alternative fixing point which is provided This is twice as far from the pivot and reduces the mechanical amplifica tion of the arm to 5 halvi
32. ion of Figure A3 it can be seen that there are consistent differences between the results for the different surface types The analysis can therefore be improved by considering the different surface types separately and the result of doing so is shown in Figure A4 Table A3 lists the regression coefficients The coefficient of determination ranges from 0 914 on asphal tic concrete surfaces to 0 987 on surface treated sec tions A 3 ALTERNATIVE PROCEDURES AND DESIGNS The simulations described so far have used one sampling procedure a Merlin of one particular size and one method of data analysis In fact the choice of these was based on other considerations and the results of other simulations The Merlin samples the road surface at a number of points and the accuracy with which roughness can be deduced clearly depends upon the quality and size of the sample It was felt that the best way of ensuring an unbiased result was to have a systematic sample with recordings taken at regular intervals The sample size 1910 593 0 04710 Asphaltic concrete Surface treated Gravel Earth 0 100 200 300 400 Merlin D mm Fig A2 Relationship between R and D Results of the Regression Analyses Roughness A A D Number of 2 R sections 0 0471 0 983 27 0 0012 47 5 0 918 27 2 8 29 9 0 914 8 3 7 37 8 0 987 5 2 5 44 0 0 967 7 3 6 59 4 0 973 7 4 4
33. is no longer 0 9 metres in this case the errors introduced are small and can be ignored 4 CALIBRATION EQUATIONS The relationships between the Merlin scale and the BI and IRI scales are given below For all road surfaces IRI 0 593 0471 D 1 42 gt D gt 312 2 4 gt IRI gt 15 9 where IRI is the roughness in terms of the International Roughness Index and is measured in metres per kilo metre and D is the roughness in terms of the Merlin scale and is measured in millimetres BI 983 47 5 D 2 42 D 312 1 270 gt gt 16 750 where BI is the roughness as measured by a fifth wheel bump integrator towed at 32 km h and is measured in millimetres per kilometre When measuring on the scale greater accuracy can be achieved by using the following relationships for different surface types Asphaltic concrete BI 574 29 9 D 3 42 lt D lt 177 1 270 lt BI lt 5 370 Surface treated 132 37 80 4 57 lt D lt 124 2 250 lt lt 4 920 Gravel 1 134 440 D 5 77 lt D 290 2 010 lt BI 12 230 Earth 2 230 59 4 D 6 B4 D 312 2 940 BI 16 750 These relationships are shown in graphical form in Figures 4 and 5 The equations were derived over the range of roughnesses shown and care should be used if extrapolating outside these ranges Roughness mm km 20 Roughness m km n 20000 15000 5000 I
34. legs additional struts are used The wheel can be D 88 mm Figure 3 Typical completed chart s JM Wo oye o gt i ns r na ra ry Neg No R114 90 5 Plate 2 Close up of probe and moving arm any type of common bicycle wheel mounted in a pair of front forks and with a tyre which has a fairly smooth tread pattern To reduce sensitivity to road surface micro texture the probe and the rear foot are both 12 mm wide and rounded in the plane of the wheel track to a radius of 100 mm The rounding also tends to keep the point of contact of the probe with the road in the same vertical line The pivot is made from a bicycle wheel hub and the arm between the pivot and the weight is stepped to avoid grounding on very rough roads The chart holder is made from metal sheet and is curved so that the chart is close to the pointer over its range of movement To protect the arm from unwanted sideways movement a guide is fixed to the side of the main beam retaining the arm close to the beam One end of this guide acts as a stop when the machine is raised by its handles The probe is attached to the moving arm by a threaded rod passing through an elongated hole a system which allows both vertical and lateral adjustment The vertical position of the probe must be set so that the pointer is close to the middle of the chart when the probe displace ment is zero or the histogram w
35. mid way between the tenth and the eleventh crosses counting in from each end of the distribution are marked on the chart below the columns It may be necessary to interpolate between column boundaries as shown by the lower mark of the TESTSECTON C7 Vp WHEEL PATH DATE 12 6 90 OPERATOR G Smith TALLY BOX 12345678910 xxxi xxx xxix 3 xix xxi 4 6 x x 19 20 ix yxp LI LT a LLL LL Ix ix Ee x ox x x x 1x L Ix example The spacing between the two marks D is then measured in millimetres and this is the roughness on the Merlin scale Road roughness in terms of the Interna tional Roughness Index or as measured by a towed fifth wheel bump integrator can then be determined using one of the equations given in Section 4 3 4 PRACTICAL DETAILS Plates 1 and 2 show the Merlin For ease of manufacture the main beam the central and rear legs the moving arm the stabiliser and the handles are all made from steel tubing of square cross section 25 x 25 mm with wall thickness of 1 5 mm Joints are welded where possible though the stabiliser and handles are fixed by bolts so that they can be removed for easier transporta tion To strengthen the joints between the main beam and the
36. mited roughness range for the latter and that the true relationships are non linear When estimating roughness for a vehicle the normal procedure is to assume that the combined roughness for the two wheel tracks can be equated to the mean of the individual tracks although this does give rise to a small error Hence in practice a minimum of two sets of Merlin observations are required The roughness of the individ ual wheel tracks can differ considerably Bearing in mind the above limitations it is normally better to calibrate an RTRRMS device at a larger number of sites than make many repeat measurements at the same site Moreover particularly if working on the BI scale these sites should have similar surfaces to those on which the RTRRMS is to be used A number of other practical points should be considered when measuring roughness or calibrating an RTRRMS and a useful guide is provided by Sayers et al 1986b As a simple cross check on performance roughness values on the Merlin scale were measured for a series of asphaltic concrete test sections on the TRRL experimen tal track Four measurements were taken on each section and the mean values are shown plotted in Figure 6 against the roughness of each section on the BI scale as measured with the Abay beam Abaynayaka 1984 The Figure also shows the Merlin BI calibration line for asphaltic concrete roads as given in equation 3 As can be seen the points lie very close to the calib
37. mp integrator is fitted on the vehicle just above the rear axle A steel hook is welded to the differential so that a E Documents Roughness Manual oad Roughness Survey Manual docPage 3 1 1 Ministry of Communication Road Roughness Roads and Highways Department Survey Manual steel wire can be connected tightly between the hook and the spring loaded pulley of the bump integrator by passing the wire through a hole made in the body of the vehicle There are many factors concerning the vehicle that will affect the readings from the bump integrator Stiffness of the vehicle suspension Condition of the shock absorbers Vehicle tyre pressure Gross vehicle weight General condition of the vehicle and engine Speed of the vehicle Driver behavior RHD 3 nos of pickups are being used since last few years for the roughness measurement 3 1 2 Bump Integrator The unit works by sensing the movement between the body of the vehicle to which the bump integrator BI is mounted and the rear axle A cable to a counter unit connects the BI unit This has two digital displays which can be switched on and off alternatively by a two way switch Each display has a clear button Picture 1 Bump Integrator Picture 2 Counter Unit Bump Integrator E Documents Roughness Manual Road Roughness Survey Manual docPage 4 1 1 Ministry of Communication Road Roughness Roads and Highways Department Survey Manual 3 1 3 Distance Measurement E
38. ng the width of the distribution Values of D read from the chart are scaled using the calibration procedure described earlier Although the spacing between the probe and the two feet is no longer 0 9 metres in this case the errors introduced are small and can be ignored 4 CALIBRATION EQUATIONS The relationships between the Merlin scale and the BI and IRI scales are given below For all road surfaces IRI 0 593 0471 D 1 42 gt D gt 312 2 4 gt IRI gt 15 9 where IRI is the roughness in terms of the International Roughness Index and is measured in metres per kilo metre and D is the roughness in terms of the Merlin scale and is measured in millimetres BI 983 47 5 D 2 42 D 312 1 270 gt gt 16 750 where BI is the roughness as measured by a fifth wheel bump integrator towed at 32 km h and is measured in millimetres per kilometre When measuring on the scale greater accuracy can be achieved by using the following relationships for different surface types Asphaltic concrete BI 574 29 9 D 3 42 lt D lt 177 1 270 lt BI lt 5 370 Surface treated 132 37 80 4 57 lt D lt 124 2 250 lt lt 4 920 Gravel 1 134 440 D 5 77 lt D 290 2 010 lt BI 12 230 Earth 2 230 59 4 D 6 B4 D 312 2 940 BI 16 750 These relationships are shown in graphical form in Figures 4 and 5 The equations were derived over the ra
39. nge of roughnesses shown and care should be used if extrapolating outside these ranges any type of common bicycle wheel mounted in a pair of front forks and with a tyre which has a fairly smooth tread pattern To reduce sensitivity to road surface micro texture the probe and the rear foot are both 12 mm wide and rounded in the plane of the wheel track to a radius of 100 mm The rounding also tends to keep the point of contact of the probe with the road in the same vertical line The pivot is made from a bicycle wheel hub and the arm between the pivot and the weight is stepped to avoid grounding on very rough roads The chart holder is made from metal sheet and is curved so that the chart is close to the pointer over its range of movement To protect the arm from unwanted sideways movement a guide is fixed to the side of the main beam retaining the arm close to the beam One end of this guide acts as a stop when the machine is raised by its handles The probe is attached to the moving arm by a threaded rod passing through an elongated hole a system which allows both vertical and lateral adjustment The vertical position of the probe must be set so that the pointer is close to the middle of the chart when the probe displace ment is zero or the histogram will not be central The lateral position of the probe has to be adjusted so that its traverse passes centrally through the line joining the bottom of the tyre and the rear foot If not
40. nternational Roughness Index Merlin D mm Bump Integrator 32km h Merlin D mm Figure 4 Calibration relationships 15 000 Gravel Bi 1134 44 00 Surface treat EN 8 574 29 9 D 20 000 E Earth Bi at 32 km h mm km 5 8 0 100 200 300 400 Merlin D mm Figure 5 Calibration relationships for Bl different surface types 5 ACCURACY OF MEASUREMENT When using the Merlin to measure roughness two considerations about accuracy have to be borne in mind The first is that the Merlin measurement for a road section is derived from a sample of observations and so is subject to a random sampling error This can be reduced by repeat observations on the same section The second is that there are systematic differences between the roughness scales which can only be reduced by repeat observations on different road sections Undulations in a road s surface can be considered as surface waves with a spectrum of spatial frequencies spectral signature The IRI Bl and Merlin scales and any RTRRMS device being calibrated all have different sensitivities to different spatial frequencies and so they correspond uniquely with each other only for surfaces with the same spectral signature In practice individual road sections have different spectral signatures though there are broad similarities especially between sections with the same surface type Hence the relationship between the scales is not unique
41. nual docPage 11 1 1 Ministry of Communication Road Roughness Roads and Highways Department Survey Manual record it on the form After that read the bump counts at every 0 5 km and record accordingly 6 DATA ENTRY The RMMS software of RHD includes standard data entry screen in which the surveyed bump counts are entered so that data is automatically stored in the RHD server Please see the RMMS user manual for detail information about roughness data entry data editing and reports E Documents Roughness Manual oad Roughness Survey Manual docPage 11 1 1 Appendix 1 CNS Farnell Distance Odometer APPENDIX 2 MERLIN MANUAL TRANSPORT AND ROAD RESEARCH LABORATORY Department of Transport RESEARCH REPORT 301 THE MERLIN LOW COST ROAD ROUGHNESS MEASURING MACHINE by MA CUNDILL Crown Copyright 1991 The work described in this report forms part of the programme carried out for the Overseas Development Administration but the views expressed are not necessarily those of the Administration Extracts from the text may be reproduced except for commercial purposes provided the source is acknowledged Overseas Unit Transport and Road Research Laboratory Crowthorne Berkshire RG11 6AU 1991 ISSN 0266 5247 CONTENTS Abstract 1 Introduction 2 Roughness measuring instruments 3 The MERLIN 3 1 Principle of operation 3 2 General description 3 3 Method of use 3 4 Practical details Calibration equations Accuracy
42. on on road profiles measured in the International Road Roughness Experi ment Sayers et al 1986a Details of this simulation are given in Appendix A 3 1 PRINCIPLE OF OPERATION The principle of operation is as follows The device has two feet and a probe which rest on the road surface along the wheel track whose roughness is to be measured The feet are 1 8 metres apart and the probe lies mid way between them see Figure 1 The device measures the vertical displacement between the road surface under the probe and the centre point of an imaginary line joining the two points where the road surface is in contact with the two feet This displacement is known as the mid chord deviation If measurements are taken at successive intervals along road then the rougher the road surface the greater the variability of the displacements By plotting the displace ments as a histogram on a chart mounted on the instru ment it is possible to measure their spread and this has been found to correlate well with road roughness as measured on standard roughness scales The concept of using the spread of mid chord deviations as a means of assessing road roughness is not new For example two roughness indices Ql and MO have been proposed by other researchers and are described by Sayers et al 1986a They are each based on the root Figure 1 Measurement of mid chord deviation ever another scale which is now being widely used is the Inte
43. operation and expensive The second class of instrument is the dynamic profile measuring device such as the TRRL high speed profil ometer Still and Jordan 1980 In these instruments surface undulations are measured with respect to a moving platform equipped with some means of compen sating for platform movement so that the true road profile can be derived This is then converted to roughness indices by automatic data processing These devices can operate at high speeds and give good quality results but they are very expensive they are not usually suitable for very rough roads and they have to be carefully main tained Finally there are the response type road roughness measuring systems RTRRMS These measure the cumulative vertical movements of a wheel or axle with respect to the chassis of a vehicle as it travels along the road In the case of a standard device such as the towed fifth wheel bump integrator Jordan and Young 1980 the response is used directly as a roughness index In other non standard devices such as the vehicle mounted BI the response is converted to a standard roughness measure by calibration The towed fifth wheel BI is expensive and needs careful operation The vehicle mounted BI however is much cheaper and can perform well as long as it is correctly used and is calibrated regularly The standard roughness scale which has been used for many years by the Overseas Unit of TRRL in its studies on v
44. ping of vehicle He should be able to connect the equipment correctly and should be well aware about the activities involved during the survey Each driver should be given training before going out for actual survey The other two members of the team will collect the data One of them will call out the reading at every 0 5 km and the other person will note down those in the standard form prepared for the survey Survey is done for the individual links of the RHD Road Network Survey must start from the starting point of the link In case if the survey is started from other than start point of the link then the actual chainage of the start point must be entered in the distance odometer during the start of the survey If in any journey more than one links are surveyed then the bump counter must be zeroed E Documents Roughness Manual Road Roughness Survey Manual docPage 9 1 1 Ministry of Communication Road Roughness Roads and Highways Department Survey Manual at the start of each link and similarly the distance odometer should also be zeroed to avoid mistakes in writing the chainage 5 1 2 Survey Methodology Prepare the equipment e Ensure that the wire of bump integrator is properly suspended with the hook below Connect the cables of the bump counters and the distance odometer properly Check that Bump counter and distance odometer displays are OK Switch on the distance odometer Turn the two way switch of bump counter to counte
45. productivity Reliable measurement of road roughness is therefore seen as an important activity in road network manage ment Several different road roughness scales have been established and a variety of roughness measuring machines have been developed However it was felt that there was a need particularly within developing coun tries for a new simple type of measuring instrument which could be used either directly to measure roughness over a limited part of the road network or for calibrating other roughness measuring equipment particularly the very widely used vehicle mounted bump integrator 2 ROUGHNESS MEASURING INSTRUMENTS Roughness measuring instruments can be grouped into three different classes The simplest in concept are the static road profile measuring devices such as the rod and level which measure surface undulations at regular intervals Unfortunately these devices are very slow in use and there can be a considerable amount of calcula tion involved in deriving roughness levels from the measurements taken Two recent devices which work on a similar principle but are semi automated are the TRRL Abay beam Abaynay aka 1984 and the modified Dipstick profiler Face Company With both of these instruments the surface undulations are measured from a static reference and data is fed directly into a microprocessor to do the necessary calculations They produce high quality results but they are relatively slow in
46. quipment For measuring the distance in kilometer a CNS Farnell Distance Odometer is being used The manual is enclosed with this instruction guide for reference Appendix 1 Picture 3 Distance Odometer 3 1 4 Merlin The MERLIN is a Machine for Evaluating Roughness using Low cost INstrumentation The Transport Research Laboratory TRL of United Kingdom devised this equipment The Merlin manual of TRL is enclosed herewith for reference Appendix 2 Merlin is simple low cost roughness measurement instrument that can be manufactured locally The instrument can be used for direct roughness measurement or for calibrating other equipment like Bump Integrator In RHD Merlin is used for calibrating purpose It consists of a metal frame 1 8 meters long with a wheel at the front a foot at the rear and a probe midway between them which rests on the road surface The probe is attached to a moving arm at the other end of which is a pointer that moves over a chart The Merlin is wheeled on the road and at every turn of the wheel the position of the pointer is recorded on the chart to build up a histogram The width of the histogram can be used to give a good estimate of roughness in terms of International Roughness Index IRI The critical measurements are the ratio of the long and short sections of the moving arm that must be 10 If there is any deviation from this ratio a correction factor must be used to obtain the correct roughness from the
47. r1 Do not switch on The team is now ready to start the survey Proceed as follows 1 Position the vehicle about 50 to 100 meters back of the starting point of the survey 2 Start driving the vehicle so that it gains the proper speed 32 km h at the start point As soon as the vehicle reaches the start point operate the distance odometer the bump integrator simultaneously Make sure that the distance odometer as well as the bump integrator is switch on 4 As the vehicle moves along the road the bump counter will display the bump readings and will continue to increase as the vehicle gets bump during running The distance odometer will continue to display the actual distance being travelling 5 When the vehicle reaches 0 5 km as will be seen in the distance odometer turn the two way switch of bump integrator to counter 2 This will cause the counter 2 to continue displaying the bump reading and counter 1 will stop recording Read the reading of counter 1 and call out so that the assistant can note down the reading in the form 6 When the vehicle reaches 1 0 km turn the two way switch towards counter 1 Now counter 1 will again continue to display cumulative readings and counter 2 will stop recording bumps Read the reading on counter 2 and call out Also record the time 7 n this way continue driving vehicle and at each 0 5 km read the corresponding bump counts and time and record those on the sheet 8 Stop the vehicle
48. ration line and while the check is by no means comprehensive it does lend strong support to the results derived from the simulation 6 DISCUSSION The reason for designing the Merlin was to provide a device which is easy to use and reasonably accurate and yet can be manufactured and maintained with the limited resources available within developing countries Experi ence indicates that it has been successful in meeting these objectives A number of the machines have been made at TRRL and shipped overseas while other units have been made overseas from drawings provided by the Laboratory To date Merlins have been used in 11 developing countries in South America Africa and Asia in six of these the equipment was made locally at current prices of typically 250 US dollars One inconvenience of the Merlin is that because of its length it is not easily transported within a vehicle A shorter machine could be used but as is shown in the Appendix this will lead to some reduction in correlation with the IRI scale Alternatively a more portable design could be considered using a structure which folds or dismantles While this is a possibility it has been avoided because of the need to retain rigidity Although its design is very simple the Merlin is able to measure displace ments to less than a millimetre and this ability could easily be compromised by unwanted flexing of the structure In recent years there has been a move towards red
49. rnational Roughness Index Sayers et al 1986a This scale which is derived from road profile data by a fairly complex mathematical procedure represents the vertical movement of a wheel with respect to a chassis in an idealised suspension system when travelling along the road at 80 km h As with the BI scale it is measured in terms of units of vertical movement of the wheel per unit length of road and is normally quoted in metres per kilometre Traditionally the BI scale is normally quoted in millimetres per kilometre 3 THE MERLIN The new instrument which has been developed is a variation of the static profile measuring device It is a manually operated instrument which is wheeled along the road and measures surface undulations at regular intervals Readings are easily taken and there is a graphical procedure for data analysis so that road roughness can be measured on a standard roughness scale without the need for complex calculation Its particular attractions for use in the developing world are that it is robust inexpensive simple to operate and easy to make and maintain The device is called MERLIN which is an acronym for a Machine for Evaluating Roughness using Low cost INstrumentation It was designed on the basis of a computer simulation of its operation on road profiles measured in the International Road Roughness Experi ment Sayers et al 1986a Details of this simulation are given in Appendix A 3 1 PRINCIPLE OF O
50. ry 1 5 metres so that the observations covered almost the entire test section In the first run the starting point was at the beginning of the test section Subsequent runs started at 0 4 0 8 and 1 2 metres from the beginning Table A2 shows the results of these simulations Values of D for each of the four runs per section are denoted as D D and D The Merlin s operation is essentially a statistical sampling of the road profile and the values of D show statistical scatter with an average coefficient of variation of eight per cent To reduce the effects of this scatter mean values of the four simulation runs are used in the analyses A plot of roughness on the IRI scale against D for each of the test sections is shown in Figure A2 As can be seen the points are a good fit to a linear regression passing close to but not through the origin Table A3 gives the As used in the IRRE NS OS Nearside Off side 1 Right Left Test Sections eee Sectn Surface Section Wheel IRI BI no type 1 code 2 track 3 m km mm km 1 AC 04 NS 4 76 3095 2 AC 04 OS 5 80 3465 3 AC 05 NS 5 68 4050 4 AC 05 OS 6 53 4390 5 AC 06 NS 6 96 4685 6 AC 06 OS 8 29 5370 7 AC 10 NS 3 29 1850 8 AC 12 OS 2 44 1270 9 ST 01 os 4 51 3280 10 ST 04 OS 5 27 3705 11 ST 05 OS 7 00 4920 12 ST 06 NS 3 11 2250 13 ST 06 OS 3 41 2725 14 GR 01 NS 3 83 2010 15 GR 05 NS 8 50 5875 16 GR 05 os 9 92 8095 17 GR 07 NS 4 11 2910 18 GR 07 o
51. s 7 04 5025 19 GR 12 NS 11 65 8545 20 GR 12 OS 14 31 12225 21 EA 01 NS 4 39 2935 22 EA 01 OS 4 72 3865 23 EA 03 NS 6 03 4315 24 EA 03 OS 8 03 8385 25 EA 06 NS 15 91 16750 26 EA 11 NS 7 78 6855 27 EA 11 os 10 78 10055 ere eee AC Asphaltic concrete ST Surface treated GR Gravel EA Earth 10 20 30 40 50 Hortzontal distance m Fig A1 Examples of test section profiles TABLE A2 Simulation Results D 7 7 Sectn Surface IRI BI mm km no type 1 m km OOnN Ons an 4 AC Asphaltic concrete ST Surface treated GR Gravel EA Earth regression coefficients together with their standard errors The coefficient of determination R is over 0 98 Hence it appears that the Merlin can be used as a fairly accurate means of measuring roughness on the IRI scale Figure shows a similar plot for roughness on the scale Once again the points can be fitted to a linear regression passing close to the origin However the fit to the line is not as good as for the IRI scale and the coefficient of determination is lower at just under 0 92 In part this was to be expected since the value was determined independently using a dynamic measuring device whereas the IRI and Merlin values were both computed from the same static profile data However this is not the full explanation and better correlation can be achieved with a Merlin of different length as described in Section A 3 1 Upon closer examinat
52. short wavelength undulations To help visualise the Merlin s response the Figure also shows the machine s length 1 8 metres on the same scale A 2 SIMULATION RESULTS Given these road profiles it was possible to carry out a computer simulation of the performance of a Merlin Neglecting the small effects due to the fact that the Merlin is not operated in a horizontal position if it is assumed that the rear foot is placed at a horizontal distance of X metres from the start of the section then the probe would be at a distance of X 0 9 metres from the start and the front foot at a distance of X 1 8 metres If the corresponding vertical distances at these points are Yo Y and then the pointer on the Merlin will be displaced from the zero position by an amount d given by M x Y 0 5 x Y Y 1 where M is the mechanical amplification provided by the moving arm usually close to 10 Placing the Merlin at successive positions along the road is simulated by using successively increasing values of X Tabulating the values of d into different 5 mm ranges corresponds to making crosses in the columns of the chart and once 200 observations have been made D can be deduced from the tabulation using the process of counting in ten observations from each end of the distribution and interpolating where necessary For each of the test sections four simulation runs were carried out In each run a Merlin reading was taken eve
53. sing just one base length the machine measures mid chord deviations without the need for rod and level the variability of the mid chord deviations is determined graphically and very little calculation is involved to determine roughness 3 2 GENERAL DESCRIPTION Figure 2 shows a sketch of the Merlin For ease of operation a wheel is used as the front leg while the rear leg is a rigid metal rod On one side of the rear leg is a shorter stabilising leg which prevents the device from falling over when taking a reading Projecting behind the main rear leg are two handles so that the device looks in some ways like a very long and slender wheelbarrow The probe is attached to a moving arm which is weighted so that the probe moves downwards either until it reaches the road surface or the arm reaches the limit of its traverse At the other end of the arm is attached a pointer which moves over the prepared data chart The Front foot with marker in contact with the road arm has a mechanical amplification of ten so that a movement of the probe of one millimetre will produce a movement of the pointer of one centimetre The chart consists of a series of columns each 5 mm wide and divided into boxes If the radius of the wheel is not uniform there will be a variation in the length of the front leg from one measure ment to the next and this will give rise to inaccuracy in the Merlin s results To overcome this a mark is painted on the
54. ss in IRI The formula is as follows IRI 0 593 0 0471D Where IRI is the roughness in terms of International Roughness Index in meters per kilometer D is the roughness in terms of Merlin scale in millimeters the base of the histogram Do the Merlin survey in 10 or 11 sites of varying roughness 4 2 3 Running Vehicles on Merlin Sites When the Merlin survey is completed the vehicles fitted with Bump Integrators are to be run on the sites to get the bump readings corresponding to the roughness of every E Documents Roughness Manual Road Roughness Survey Manual docPage 7 1 1 Ministry of Communication Road Roughness Roads and Highways Department Survey Manual site Bump readings are to be taken for the distance between the paint marks put during the survey Be sure that the vehicle is run at a speed of 20 mph 32 km h To attain the speed start running the vehicle from at least 100m back of the starting mark When the vehicle reaches the start mark make the Bump Integrator on and stop the Bump Integrator at the end mark Note down the bump count thus obtained Do this exercise several times to get at least four consistent readings and average them This average bump reading represents the bump count against the roughness obtained by Merlin In this way run other survey vehicles in each site and get the corresponding bump counts against the roughness 4 2 4 Doing the Regression Analysis On next page is a worksheet showing
55. the data of nine Merlin sites The D value of each Merlin has been corrected using the correction factors of Merlins Picture 5 Merlin Worksheet pe LS pese Ha hhc dl Al ell i i Bl hl li Kel ia ll eo ad Ro Ra K o Ei OO o Bi Go d a Mises B Ro NL RU Dod RR Ra i Rod Ea esac Mr wi The regression analysis chart with the line fit plot and the formula obtained for the Roughness and Bump count relationship is shown below E Documents Roughness Manual Road Roughness Survey Manual docPage 8 1 1 Ministry of Communication Road Roughness Roads and Highways Department Survey Manual Picture 6 Regression analysis chart Calibration of Bump Integrator 2000 Vehicle Codez4 Reg no 02 0010 y 0 0144x 0 6242 R 0 987 Bump Reading 5 DOING THE ACTUAL ROUGHNESS SURVEY The vehicles calibrated by the above method are then used for actual survey The intention is to collect the bump counts at every 0 5 km of roads These bump counts are utilized to get the roughness of road by using the formula obtained from the above regression analysis 5 1 1 Survey team The survey team for vehicle consists of e Driver e Assistant Engineer e Assistant The driver plays the important role during the survey He keeps the vehicle well maintained During survey he drives carefully to maintain the speed avoids abrupt jum
56. this is to some extent explained by the fact that the BI value was determined by independent measurement The use of a one metre Merlin is an attractive concept since it would be considerably more portable than the 1 8 metre version However it would be a much poorer predictor of IRI and in practice it would be necessary to distinguish between the different surface types to reduce some of the uncertainty The underlying reason for the results of this analysis can be explained by considering the frequency sensitivities of the Merlin and the IRI and BI scales The Merlin has a fundamental frequency response to surface waves of wavelength equal to its own base length while the IRI and scales are particularly sensitive to surface waves which would stimulate the natural vibrations of a vehicle wheel at about 10 Hz and a chassis at about 1 Hz At 80 km h the speed used for the IRI scale the natural vibration of the wheel would be stimulated by surface waves of around 2 2 metres and the chassis by waves of around 22 metres At 32 km h the speed used for the scale the equivalent surface wavelengths are 0 9 metres and 9 metres respectively Hence it appears that the correlation analysis has selected Merlin lengths such that the wavelength of the fundamental frequency is close to the wavelength of the surface waves which would stimulate the natural vibration of the wheel A 3 2 Measurement of data spread Finally the choice of me
57. thod for determining the data spread should be described Measuring the limits for a certain central percentage of the data points is an BI at 32 km h mm km 5 000 Figure A3 200 300 Merlin D mm Relationship between Bl and D Surface treated Bi mm km 20 000 2 15 000 10 000 0 100 200 300 400 Merlin D mm Figure A4 Relationship between Bl and D for different surfaces Coefficient of Determination R 2 Figure AS Roughness measuring accuracy for Merins of different length attractively simple procedure in the field and requires a minimum of calculation To decide what percentage would give the best answers the performance of a Merlin over the test sections was again simulated This time the machine length was fixed at 1 8 metres and the rough ness was measured on the IRI scale Linear regressions were carried out between D values derived using different data percentages and roughness Table A4 shows the resulting values of R from which it can be seen that of the values tested 90 per cent which corresponds to counting in 10 crosses from each end of the distribution appeared to be the best choice TABLE A4 Effect of Data Limits on Correlation Percentage Count from edge R of data of distribution 95 5 0 932 90 10 0 983 85 15 0 966 80 20 0 923 Printed in the United Kingdom for HMSO DdK50120 8 91 C10 G2516 10170 S
58. tions along the road is simulated by using successively increasing values of X Tabulating the values of d into different 5 mm ranges corresponds to making crosses in the columns of the chart and once 200 observations have been made D can be deduced from the tabulation using the process of counting in ten observations from each end of the distribution and interpolating where necessary For each of the test sections four simulation runs were carried out In each run a Merlin reading was taken every 1 5 metres so that the observations covered almost the entire test section In the first run the starting point was at the beginning of the test section Subsequent runs started at 0 4 0 8 and 1 2 metres from the beginning Table A2 shows the results of these simulations Values of D for each of the four runs per section are denoted as D D and D The Merlin s operation is essentially a statistical sampling of the road profile and the values of D show statistical scatter with an average coefficient of variation of eight per cent To reduce the effects of this scatter mean values of the four simulation runs are used in the analyses A plot of roughness on the IRI scale against D for each of the test sections is shown in Figure A2 As can be seen the points are a good fit to a linear regression passing close to but not through the origin Table A3 gives the JORDAN G and YOUNG J C 1980 Developments in the calibration an
59. ucing the number of different roughness scales in use and standardising on the International Roughness Index However the Merlin scale does have the advantage of being easy to visualise and although Merlin readings can be converted easily to IRI values in some cases this conversion is unnecessary and direct use of the Merlin scale should be considered 7 ACKNOWLEDGEMENTS This work forms part of the programme of research of the Overseas Unit Head J S Yerrell of the Transport and Road Research Laboratory UK 8 REFERENCES ABAYNAYAKA S W 1984 Calibrating and standardis ing road roughness measurements made with response type instruments In ENPC International Conference on Roads and Development Paris May 1984 pp13 18 Presses de l ecole nationale des ponts et chaussees Paris CHESHER A and HARRISON R 1987 Vehicle operating costs evidence from developing countries John Hopkins University Press Baltimore and London CRRI 1982 Road user cost study in India final report Central Road Research Institute New Delhi FACE COMPANY The Edward W Face Company Inc Norfolk Virginia GILLESPIE T D 1986 Developments in road rough ness measurement and calibration procedures In ARRB Proc 13th ARRB 5th REAAA Conf 13 1 pp 91 112 Australian Road Research Board Vermont South HIDE H 1982 Vehicle operating costs in the Carib bean results of a survey of vehicle operators TRRL Laboratory Report
60. ve to be carefully main tained Finally there are the response type road roughness measuring systems RTRRMS These measure the cumulative vertical movements of a wheel or axle with respect to the chassis of a vehicle as it travels along the road In the case of a standard device such as the towed fifth wheel bump integrator Jordan and Young 1980 the response is used directly as a roughness index In other non standard devices such as the vehicle mounted BI the response is converted to a standard roughness measure by calibration The towed fifth wheel BI is expensive and needs careful operation The vehicle mounted BI however is much cheaper and can perform well as long as it is correctly used and is calibrated regularly The standard roughness scale which has been used for many years by the Overseas Unit of TRRL in its studies on vehicle operating costs and pavement deterioration is the output of the fifth wheel BI towed at 32 km h How THE MERLIN LOW COST ROAD ROUGHNESS MEASURING MACHINE ABSTRACT The roughness of a road s surface is an important measure of road condition and a key factor in determining vehicle operating costs on poor quality surfaces This report describes a simple roughness measuring machine which has been designed especially for use in developing countries It is called MERLIN a Machine for Evaluating Roughness using Low cost INstrumentation The device can be used either for direct measurement
61. when the survey for a link is completed 9 Please record the closing distance exactly from the distance odometer The final section which is less than 0 5 km is called the closing fraction At this point please note that if the link you surveyed is a narrow road single lane say 4m then you do not need to survey back to the start of the link Otherwise you will have to collect bump readings of the other side of the road by doing a back survey from the end of the link to the start point During the back journey record the bump counts against the corresponding chainage already recorded in the sheet which means that you have to start writing bump counts on the sheet from bottom to top During back journey repeat step 1 and 2 above Drive to the distance amounting to the closing fraction and clear the distance odometer Read the bump reading and E Documents Roughness Manual oad Roughness Survey Manual docPage 10 1 1 Ministry of Communication Road Roughness Roads and Highways Department Survey Manual record it on the form After that read the bump counts at every 0 5 km and record accordingly 6 DATA ENTRY The RMMS software of RHD includes standard data entry screen in which the surveyed bump counts are entered so that data is automatically stored in the RHD server Please see the RMMS user manual for detail information about roughness data entry data editing and reports E Documents Roughness Manual oad Roughness Survey Ma

Roughness Manual - Roads and Highways Department

Contents

Download Pdf Manuals

Related Search

Related Contents