The Tractive Performance of Rubber Tracks and a Tractor Driving Wheel Tyre as Influenced by Design Parameters

Transcription

1 J. agric. Engng Res. (1994) 59, The Tractive Performance of Rbber s and a Tractor Driving Wheel Tyre as Inflenced by Design Parameters J. A. Okello;* M. J. Dwyer;t F. B. Cottrell* *Soil Science Grop, Mechanical Engineering Division, Silsoe Research Institte, Wrest Park, Silsoe, Bedford, MK45 4HS, UK t Ministry of Agricltre, Food and Fisheries (MAFF), Nobel Hose, 17 Smith Sqare, London SWlP 3JR (Received 31 Agst 1993; accepted in revised form 21 December 1993) The tractive performance of an eperimental rbber track nit was compared with that of a 16 9 R34 the overall performance of agricltral tractors wold clearly be considerably improved if they cold be tractor driving wheel tyre, in 14 different field fitted with a grond-driven system which combined conditions, sing a single wheel tester. In 1989, the track nit was fitted with a thinner, fleible, friction-driven rbber track having less aggressive lgs than that of the tyre. Tests were carried ot with vertical loads of kn and kn for both the track and the tyre in eight different field conditions. the good tractive capability and low grond pressre of the steel track with the non-abrasive featres and higher speed, road-going capability of the pnematic tyre. This led to the consideration of the potential for se of rbber tracks reviewed by Clshaw.1 2 By the mid-l 980s, very little was pblished on the The rolling resistance of the track was fond to be performance of rbber tracks. The Bonmartini higher than that of the tyre becase of the internal power losses within the track and losses between the driving sprocket and the track. The coefficient of traction prodced by the track was higher than that of the tyre, bt the tractive efficiency was lower, de to its higher rolling resistance. The coefficient of traction at which maimm tractive efficiency was achieved with the track was always mch higher and the corresponding slip was always mch lower than those achieved with the tyre. The significance of this is that a rbber-tracked vehicle cold be mch lighter than a wheeled tractor of the same power and still achieve the same tractive performance. In 19, the same tests were repeated with two thicker, positively driven rbber tracks having different tread patterns, and with the same tyre in si different field conditions. The reslts of the comparison confirmed those of the previos investigation. The tread patterns had no pnematic track was the first to be sccessflly tested against a pnematic tyre and steel track by Taylor and Brt.3 However, it cold not be adopted commercially, presmably becase of its compleity and cost of manfactre. Research work on rbber tracks at Silsoe Research Institte attempted initially to se a friction drive to the rbber track from a pnematic tyre. 4 5 This was an improved design in which a rbber belt ran on two dal pairs of pnematic tyres, in tandem. This proved impractical becase, althogh there was provision for lateral location which prevented the belt from slipping off sideways, tangential slip still occrred between the belt and the tyres at high torqe. This clearly demonstrated that it was essential to have a design with a positively driven track to prevent slip between. the track and the drive sprocket if very high track tension is to be avoided. statistically significant effects on their tractive The first satisfactory comparative data between the performance. tractive performance of a tyre and rbber track were obtained sing a small, positively driven, rbbertracked dmper and a conventional two-wheel drive 1- Introdction tractor having the same weight on its driving ale.6 Reslts showed that the tracked vehicle prodced For the last 50 years, the choice of tractive devices higher drawbar pll with less slip, bt had a higher sed on agricltral tractors for generating high rolling resistance, reslting in similar maimm tractive forces on soft soil srfaces, necessary for tractive efficiency to the wheeled vehicle. plling ploghs and cltivation implements, has been The most recent developments in the commercial between pnematic tyres and steel tracks. However, application of rbber tracks to agricltral vehicles 33 OOZJ.8634/94/ $08.00/ Silsoe Research Institte

2 34 J. A. OKELLO ET AL. Notation A. B constants or coefficients C, coefficient of rolling resistance C, coefficient of traction R wheel rolling radis or track sprocket radis, m V vehicle forward speed, m/s s wheel or track slip 1'/1 tractive efficiency w wheel or track rotational speed, rad/s have been the introdction of the flly tracked Challenger tractor' and the Marshal tractor. 8 The rbber tracks on the Challenger are frictiondriven from a solid, rbber-faced drive wheel with a pnematic-tyred front idler in which slip between the driving wheel and the track is avoided by setting a very high track tension. The Marshall, on the other hand, is fitted with 510 mm wide elastomeric polymer tracks positively driven by a steel cage wheel sprocket, at the rear, engaging with a central row of lgs on the inside of each track. The tractive performance of these tractors and wheeled tractors of corresponding similar weights and power in tilled, stbble and firm soils showed better tractive performance from the rbber-tracked vehicles with consistently higher pll, greater efficiency and lower or comparable rolling resistance compared to the wheeled tractors. This work has shown that replacement of pnematic tyres with rbber tracks on agricltral vehicles can potentially redce soil compaction, redce top soil shear damage and increase tractive performance. This helps to perserve a good stable soil strctre, withot which water rn-off and erosion can lead to transport of polltants into water corces and biological activity will release greenhose gasses into the atmosphere. In fact, the reslts of earlier work showed that, in general, tracks, whether steel or rbber, give a higher coefficient of traction than tyres, bt particlarly if positively driven, may have a higher rolling resistance. In that case, they wold have a lower tractive efficiency bt this occrs at a higher vale of coefficient of traction. In this paper, the reslts of the comparison of the tractive performance of a rbber track and a tyre, sing two types of drive system to the track, and the comparison of two tracks with different tread patterns are described. Reslts of the work described in this paper led to the design and constrction of niqe rbber track nits incorporating their own sspension systems to replace the resilience normally provided by pnematic tyres.9 A conventional, for-wheel drive tractor was fitted with these nits to replace the rear wheels. The comparison of the rbber-tracked tractor verss conventional wheeled tractors of the same power in normal field conditions has shown the speriority in tractive performance (20 to '30% more) and many other attractive featres of rbber tracks Eperimental eqnipment Two types of rbber tracks, made by two different companies, were sed in the eperiment which was performed over two consective years 1989 and 19. The first one, made by a French company, the Colmant-Cvelier, primarily for over-snow vehicles, which has also been developed as a half-track system for agricltral tractors nder the name Kitco, was tested in In this eperimental design, gide horns were sed to provide a positive drive from a specially designed alminim sprocket. The drive system, shown in Fig. 1, consisted of a series of rbber gide horns rnning on the inside, on either side of the rbber track, engaging with the teeth of a specially designed drive sprocket at the rear of a special track nit to prevent the track slipping sideways, and a high track tension to redce damage to the gide horns and stop tangential slip at high torqe. The important track dimensions are shown in Table 1. The second type consisted of two tracks both made by the Japanese Bridgestone company, primarily for earthmoving eqipment and were tested in 19. They are reinforced longitdinally with steel wire and transversely with steel plates which engage with the teeth of the driving sprocket. These two tracks were designed to be positively driven by a steel cage wheel sprocket, at the rear of the track nit, engaging with a central row of rbber-covered metal lgs and holes rnning on the inside of the track as shown fitted into the single wheel tester in Fig. 2. One of the tracks had straight transverse lgs and the other had lgs set at an angle as shown in Fig. 3. The main track dimensions are shown in Table 2. All these three tracks were fitted into the single wheel tester, one at a time, as shown in Fig. 2 and their tractive performance compared with that of a conventional 16 9 R34 radial ply tractor driving wheel tyre with Rl tread pattern. They were all tested with the same vertical loads of kn and kn. The tyre was inflated to 100 kpa for the lower load and 130 kpa for the higher load. It was accepted that the first arrangement was likely to lead to high rolling

3 PERFORMANCE OF RUBBER TRACKS 35 Fig. 1. Rbber track nit (Colmant-Cvelier track) Table 1 dimensions in mm (Colmant-Cvelier track) width length thickness Lg height 375 Sprocket and idler diameter 5040 Sprocket to idler centre distance 14 Roller diameter Fig. 2. Single wheel tester with track nit (fitted with Bridgestone Rbber ).

4 36 J. A. OKELLO ET AL. lgs width length Angled lgs Fig. 3. Bridgestone rbber track tread patterns. Table 2 dimensions in mm (Bridgestone tracks) thickness Lg height Lg width at tip Lg pitch Sprocket and idler diameter Sprocket to idler centre distance Roller diameter lgs Angled lgs resistance bt the chevron lgged track, made by Colmant-Cvelier, was qite thin in comparison to that prodced by Bridgestone which was only for testing in the second drive system. J_ Eperimental procedre The single wheel tester 10 is fitted with two independent hydrostatic transmissions, one of which drives the test wheel or track and the other drives the main vehicle. A traction test is carried ot by maintaining the test wheel or track rotational speed approimately constant while gradally redcing the forward speed of the vehicle. This cases the torqe inpt to the test wheel or track, the forward thrst prodced and the slip to increase. These parameters are recorded continosly in real time by means of an on-line digital compter data logger hosed in the instrment test compartment in the tractor cab. The rolling resistance of the tyre or track was first measred by driving the tester forward, with zero torqe in the driveline to the test carriage, and measring the towing force reqired. This force was divided by the vertical load to obtain the coefficient of rolling resistance. Wheel or track slip was obtained from the rotational speed of the drive to the test carriage and the forward speed of the vehicle measred by a small trailing track nit shown in Fig. 2. Zero slip was said to occr when jst sfficient torqe was applied to overcome the rolling resistance at zero forward thrst. Slip was defined as v s=l-- (1) Rw where s = slip V = vehicle forward speed R = wheel rolling radis or track sprocket radis w = wheel or track rotational speed and tractive efficiency was defined as C,(1-s) q =, c I + c ' where 111 =tractive efficiency C, =coefficient of traction (net forward thrst/vertical load) C, = coefficient of rolling resistance Each test for the tyre or track for a given vertical load consisted of si separate rns in different locations in the field to minimize the effect of variations in soil conditions within the test area. Tests were started with a no-slip speed of approimately 4 km/h and data from the si rns were combined to form a single set of test reslts. Soil conditions were defined by particle size classification, plastic limit and moistre content all of which were assessed in the laboratory and the average cone penetrometer resistance was measred in sit throgh the top 225 mm depth. Soil cohesion, friction angle and plate sinkage parameters were measred in the field sing the Soil Properties Rig (SPRIG) described by Mahmod and Dwyer.11 Since the gide horns on the inside of the Colman! Cvelier rbber track were sed to provide a positive drive from a specially designed alminim sprocket described by Dwyer et al, 12 it was fond necessary to condct a frther series of measrements of rolling resistance on a concrete test track to investigate factors affecting the internal power losses in this rbber track nit. 4. Field conditions (2) Tests were carried ot with the Colmant-Cvelier track in eight field conditions described in Table 3 and

5 PERFORMANCE OF RUBBER TRACKS 37 Table 3 Field conditions (Colmant Cvelier track, 1989) Cone Plastic Moistre inde, Field no Month Srface Soil ty pe limit, o/o content, % kpa 1 Janary Stbble Clay March Grass SCL May Grass Clay May Disced Clay May Cltivated Clay 27 6 Agst Stbble SCL September Stbble SCL September Ploghed SCL SCL Sandy clay loam. Fields 3, 4 and 5 and fields 7 and 8 were adjacent to each other. Table 4 Field conditions (Bridgestone tracks, 19) Field no Month Srface Cone Soil Plastic Moistre inde, ty pe limit, % content, % kpa 1 Janary 2 Janary 3 Jne 4 Jly 5 Jly 6 September Ploghed and rolled Ploghed and rolled Grass Disced Stbble Ploghed SCL SCL Clay Clay SCL SCL SCL Sandy clay loam *Too sandy to record with the two Bridgestone tracks in si field conditions described in Table 4. Moistre content and cone penetrometer resistance were not measred in field 5 becase a severe change in weather developed soon after completing traction tests which forced soil measrements to be abandoned. Cone penetrometer resistance in fields 6 and 7 were too high to record with the hand-operated cone penetrometer available and moistre content in fields 7 and 8 was too low to have any relevance. 5. Reslts Some eamples of the relationship between coefficient of traction and slip and between tractive efficiency and coefficient of traction for the tyre and the Colmant-Cvelier track respectively with a vertical load of kn, in field 8 (Table 3), are shown in Figs 4 to 7. Similar eamples for the same tyre and the Bridgstone track respectively with the same load (16 68 kn), in field 5 (Table 4), are shown in Figs 8 to 11. Some of the eperimental data have negative slip bt corresponding positive pll. This is de to the random sampling techniqe of data employed by the single wheel tester instrmentation in real time. When c g 0 6 'O E o IE Slip Fig. 4. Coefficient of traction verss sl ip for the tyre in field 8 (Vertical load, 16 68kN).

6 38 ). A. OKELLO ET AL. l O c 0.., g 'O E "' g "' 70 ;; c " E " > a,g 20 '""--,.., ---'--,..,..,...,_ 00 O I Slip Coefficient of traction Fig. 5. Coefficient of traction verss slip for the Colmant Cvelier rbber track in field 8. (Vertical load, kn) Fig. 7. Tractive efficiency verss coefficient of traction fo r the Colmant-Cvelier rbber track in field 8 (Vertical load, 16 68kN). this sampling process takes place at low vales of slip, some data will fall below or above zero slip and hence the negative data. However, some data have negative slip and corresponding negative pll vales. This occrs dring the initial set p for no slip condition (described by Koolen and Kipers 13) in which the wheel or track speed can be momentarily less than the single wheel tester speed reslting in wheel or track skid. Crves were fitted to all the coefficient of traction verss slip relationships, by a least sqares method, assming the relationship to be of the form C, =A[ 1- (1- e-< 18l) J (3) as derived originally by Bekker 14 and modified by Janosi and Hanamoto 15 where A and B are coefficients derived from the crve fitting rotine. Crves for the tractive efficiency and coefficient of traction were derived from Eqns (2) and (3) and the measred vale of rolling resistance. The effects of the two vertical loads on maimm tractive efficiency, maimm coefficient of traction and slip at maimm tractive efficiency for either the tyre or the tracks tested in the same fields were fond to be insignificant. That is, the differences in the reslts obtained for a given track or the tyre tested in a given field condition with the two vertical loads were "' 70 ;; 60 c IE 50 " " >.B 40.= c 'O E IE " Coefficient of traction Fig. 6. Tractive efficiency verss coefficient of traction for the tyre in field 8 (Vertical load, kn) Slip Fig. 8. Coefficient of traction verss slip for the tyre in field 5 (Vertical load, kn).

7 PERFORMANCE OF RUBBER TRACKS 39 l O , "f1!\ 'C E o "' O O O lf.(1< O l Slip Coefficient of traction Fig. 9. Coefficient of traction verss slip for the Bridgestone rbber track in field 5 (Vertical load, kn). Fig. 11. Tractive efficiency verss coe fficient traction for the Bridgestone rbber track in field 5 (Vertical load, kn). fond to be statistically insignificant. The important parameters directly related to data obtained from the fields are smmarized in Tables 5 to 14 and they clearly show minor differences for the two load conditions. For this reason and also becase there is too mch data available, only a few eamples from the several field reslts were taken from fields 5 and 8 for jst one load condition and presented in the form of fitted crves (i.e. coefficient of traction verss slip and the corresponding tractive efficiency verss coefficient of traction (Figs 4 to 11) at kn load) for prposes of illstration. The reslts of Fig. 6 are "" 70 "' " 60 le 50 >.,, O O O l Coefficient of trac1ion Fig. 10. Tractive efficiency verss coefficient of traction for the tyre in field 5 (Vertical loa d, kn). derived directly from Fig. 4 sing Eqns (2) and (3) and similarly Figs 5 and 7, 8 and 10, and 9 and 11 correspond to each other. However, as mentioned above, measred vales of coefficient of rolling resistance and the vales of maimm coefficient of traction, maimm tractive efficiency and coefficient of traction and slip at maimm tractive efficiency for the Colman! Cvelier track obtained from the fitted crves are shown in Tables 5, 6, 7, 8 and 9 respectively. The vales of the same parameters for the Bridgestone tracks are shown in Tables 10, 11, 12, 13 and 14 respectively. Reslts of the measrement of coefficient of rolling resistance of the Colmant-Cvelier track on concrete, with kn vertical load, are shown in Table 15. Table S Coefficient of rolling resistance (Colmant-Cvelier track) Vertical load 16 68kN kn Field Tyre Tyre Tra ck Mean (). 19

8 40 J. A. OKELLO ET AL. Table 6 Table 7 Maimm coefficient or traction (Colmant-Cvelier Maimm tractive efficiency, % (Colmant-Cvelier track) track) Vertical Vertical load 16 68kN kn load!6 68kN 19 62kN Field Tyre Tyre Field Tyre Tyre I Mean Mean Table 8 Table 9 Coefficient of traction at maimm tractive efficiency (Colmant-Cvelier track) Slip at maimm tractive efficiency, (Colmant-Cvelier track) Vertical load 16 68kN kn Vertical load 16 68kN kn Field Tyre Tyre Field Tyre Tyre l Mean Mean Table 10 Coefficient or rolling resistance (Bridgestone tracks) Vertical load 16 68kN kn Field no. Tyre lgs Angled lgs Tyre lgs Angled lgs Mean

9 PERFORMANCE OF RUBBER TRACKS 41 Table 11 Maimm coefficient of traction (Bridgestone tracks) Vertical load 16 68kN 19 62kN Field no. Tyre lgs Angled lgs Tyre lgs Angled lgs Mean Table 12 Maimm tractive etliciency, 0/o (Bridgestone tracks) Vertical load 16 68kN 19 62kN Field no. Tyre lgs Angled lgs Tyre lgs Angled lgs Mean Table 13 Coefficient of traction at maimm tractive efficiency (Bridgetone tracks) Vertical load 16 68kN 19 62kN Field no. Tyre lgs Angled lgs Tyre lgs Angled lgs Mean

10 42 J. A. OKELLO ET AL. Table 14 Slip at maimm tractive efficiency, 0/o (Bridgestone tracks) Vertical load 16 68kN kn Field no. Tyre lgs Angled lgs Tyre lgs Angled lgs Mean Table 15 Coefficient of rolling resistance of the Colmant-Cvelier track on concrete Speed, km/h Standard arrangement O 21} Coefficient of With drive to sprocket disconnected rolling resistance, C, (Tension = 30 kn) tension, KN Coefficient of rolling resistance, C, at 5 km/h 18 } Standard 0.15 arrangement 6. Discssion All the reslts were analysed statistically to determine significant differences. Reslts showed that differences between the Colman! Cvelier track and the tyre were all significant, for maimm tractive efficiency at 95% probability and for all other parameters at 99%. As eplained above, the effects of the two vertical loads on the important parameters were fond to be insignificant. The rolling resistance of the Colman! Cvelier track was more than twice as high as that of the tyre in all fields ecept field 8 (Table 3), ploghed bean stbble. The rolling resistance of the tyre tested with both the Colman! Cvelier and Bridgestone tracks varied considerably depending on the soil conditions indicating that it was mainly cased by tyre/soil interaction. The rolling resistance of the rbber tracks, on the other hand, was largely independent of the soil conditions. The rolling resistance of the tracks when carrying a vertical load of kn was higher than when carrying kn with 99% probability. The rts formed by the tracks were shallower than those cased by the tyre in all cases. Ths, the major part of the rolling resistance of the tracks mst have been de to internal power losses and therefore independent of the vertical load. These losses cold be owing to friction at the teeth of the driving sprocket, hysteresis in the rbber as it fleed arond the sprocket and idler pls friction in the bearings of the sprocket, idler and rollers. This was confirmed by tests carried ot on the Colman! Cvelier on concrete. These tests showed that rolling resistance increased with speed and tension. These power losses were fond to be higher than those measred with the Bridgestone tracks. The maimm coefficient of traction of the rbber tracks was higher than that of the tyre with 99% probability. In fields 3 and 4, however, (Table 4) for the Bridgestone tracks both of which had a looser srface overlying firmer soil, the tyre prodced higher maimm coefficients of traction than the tracks. This cold have been de to the higher contact pressre or more aggressive lgs on the tyre which penetrated the srface cover digging down into the stronger soil below. -slip was, however, always less than the wheel slip over the fll range of coefficient of traction. This is the reslt of the larger, longer area of grond contact nder the track. The maimm tractive efficiency of the tyre was

11 PERFORMANCE OF RUBBER TRACKS 43 always higher than that of the rbber tracks ecept for the Colman! Cvelier track in field 8 (Table 3) where the rolling resistance of the tyre was higher becase of the loose soil srface. The coefficient of traction and maimm tractive efficiency was higher for the rbber tracks than for the tyre, with 99% probability and the corresponding slip was always lower. This confirms the potential of rbber tracks to provide better tractive performance than a wheeled tractor either over a wider range of drawbar pll with a tractor of the same weight or over the same range with a mch lighter tractor. 7. Conclsions 1. The rbber tracks had higher rolling resistances than the tractor driving wheel tyre. This appeared to be de to internal power losses in the track nit. However, the rolling resistances de to soil compaction and rtting appeared to be mch less with the rbber tracks than with the tyre. 2. The rbber tracks prodced higher vales of coefficient of traction than the tyre. 3. The tractive efficiencies of the rbber tracks were lower than that of the tyre becase of their higher rolling resistances. 4. The coefficient of traction of the tracks at maimm tractive efficiency was consistently higher than that of the tyre and the slip at maimm tractive efficiency was consistently lower. 5. The differences between the tread patterns of the rbber tracks had no statistically significant effect on their tractive performance. 6. These reslts show that rbber tracks have great potential to provide better tractive performance than a wheeled tractor either over a wide range of drawbar pll with a tractor of the same weight or over the same range with a mch lighter tractor. This means that soil strctral damage, especially sbsoil compaction de to vehicle weight, can potentially be redced or avoided throgh design. References 1 Clshaw, D. Rbber tracks for agricltre. Divisional Note, DN 1345, AFRC Engineering Clshaw, D. Rbber tracks for traction. Jornal of Terramechanics 1988, 25: Taylor, J. H.; Brt, E. C. and tyre performance in agricltral soils. Transactions of the ASAE 1975, 18: Mickelson, P. I. Rbber tracks for machinery and vehicles in forestry and agricltre. Proc. 8th International Conference of!stys, Cambridge, Clshaw, D.; Dawson, J. R. The performance of a simple rbber track for an agricltral vehicle. Divisional Note, DN 1382, AFRC Engineering Silsoe Clshaw, D. Tractive performance of a small rbber tracked vehicle. Divisional Note, DN 1389, AFRC Engineering Silsoe Esch, J. H.; Bashford, L. L.; Von Bargen, K.; Ekstrom, R. E. Tractive performance comparisons between a rbber belt track and a for-wheel drive tractor. Transactions of the ASAE 19, 33( 4) 8 Dwyer, M. J.; Okello, J. A.; Scarlett, A. J. A theoretical and eperimental investigation of rbber tracks for agricltre. ISTVS/FISIT A Conference, London, Reed, J. N.; Scarlett, A. J. Design and evalation of a rbber-tracked agricltral tractor. 5th Eropean Conference of the!stys, Hngary, 1991, 2 '0 Billington, W. P. The NIAE Mk II Single Wheel Tester Jornal of Agricltral Engineering Research 1973, 18: Mahmod, M.; Dwyer, M. J. The effect of tyre-soil interface shape on the prediction of vehicle off-road performance. Proceedings of the 5th Eropean Conference of the ISTVS, Bdapest, 1991 Dwyer, M. J.; Okello, J. A.; Cottrel, F. B. A comparison of the tractive performance of a rbber track and a tractor driving wheel tyre. Proceedings of the 10th International Conference of the ISTVS, Kobe, Koolen, A. J.; Kipers, H. Agricltral soil mechanics. Springer-Verlag, Bekker, M. G. Off-the-road locomotion. Ann Arbor: University of Michigan Press, Janosi, Z.; Hanamoto, B. Analytical determination of drawbar pll as a fnction of slip for tracked vehicles in deformable soils. Proceedings of the 1st ISTVS Conference, Trin, 1961