Journal of Sports Sciences A new method to measure

This article was downloaded by: [University of Lausanne] On: 16 August 2010 Access details: Access Details: [subscription number 786945810] Publisher Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 3741 Mortimer Street, London W1T 3JH, UK

Journal of Sports Sciences

Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t713721847

A new method to measure rolling resistance in treadmill cycling Yves Henchoza; Giacomo Crivellia; Fabio Borrania; Grégoire P. Milleta a Institute of Sports Sciences, University of Lausanne, Lausanne, Switzerland First published on: 04 August 2010

To cite this Article Henchoz, Yves , Crivelli, Giacomo , Borrani, Fabio and Millet, Grégoire P.(2010) 'A new method to

measure rolling resistance in treadmill cycling', Journal of Sports Sciences,, First published on: 04 August 2010 (iFirst) To link to this Article: DOI: 10.1080/02640414.2010.498483 URL: http://dx.doi.org/10.1080/02640414.2010.498483

PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf This article may be used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

Journal of Sports Sciences, 2010; 1–4, iFirst article

A new method to measure rolling resistance in treadmill cycling

YVES HENCHOZ, GIACOMO CRIVELLI, FABIO BORRANI, & GRE´GOIRE P. MILLET Institute of Sports Sciences, University of Lausanne, Lausanne, Switzerland

Downloaded By: [University of Lausanne] At: 10:35 16 August 2010

(Accepted 1 June 2010)

Abstract The purpose of this study was to evaluate a new method of measuring rolling resistance in treadmill cycling and to establish its sensitivity and reproducibility. One participant was asked to keep a bicycle in equilibrium on a treadmill without pedalling at a constant speed of 5.56 m s71, which was held in place in the front by a dynamometer. For each condition, the method consisted of 11 measurements of the force required to hold the cycle at different treadmill slopes (0–10%, increment 1%). The coefficient of rolling resistance was calculated based on the forces applied to the bicycle in equilibrium. To test the sensitivity of the method, the bicycle was successively equipped with three tyre types (700 6 28, 700 6 23, 700 6 22) and inflation pressure was set at 150, 300, 600, 900, and 1100 kPa. To test the reproducibility of the method, a second experimenter repeated all measurements done with the 700 6 23 tyres. The method was sensitive enough to detect an effect of both tyre type and inflation pressure (P 5 0.001: two-way ANOVA). The measurement of the coefficient of rolling resistance by two separate experimenters resulted in a small bias of 0.00029 (95% CI, 70.00011 to 0.00068). In conclusion, the new method is sensitive and reliable, as well as being simple and affordable.

Keywords: Rolling resistance, treadmill cycling, biomechanics

Introduction At low cycling speeds, rolling resistance exceeds air resistance and constitutes a significant parameter of performance (Pugh, 1974). Several measurement methods have been proposed. Some authors measured rolling resistance by establishing the relationship between the total force opposing motion and the squared cycling speed, either in wind tunnels (Davies, 1980) or in the field (Pugh, 1974). Based on previous experimental data, a theoretical model calculated rolling resistance according to tyre inflation (Whitt, 1971). Another technique consisted in towing a cyclist by a vehicle on a flat track at constant speed and measuring the total force opposing motion using a dynamometer (Capelli et al., 1993; di Prampero, Cortili, Mognoni, & Saibene, 1979). Finally, a simplified deceleration method was recently developed, where the coefficient of rolling resistance was derived from the equation for resistive forces opposing motion (Candau et al., 1999). Coleman and colleagues recently highlighted the advantages of treadmill cycling over ergometers when assessing individuals (Coleman, Wiles, Davison, Smith, & Swaine, 2007). They demonstrated that treadmill cycling can be used as an accurate and

functional system. A limitation of the method, however, was the extrapolation of the rolling resistance value instead of a direct measurement. All the methods mentioned above require either an expensive and substantial system, or do not enable rolling resistance to be measured on a treadmill. The aims of this study were to present a new method to measure rolling resistance in treadmill cycling and to establish its sensitivity and reproducibility.

Methods Description of the method All measurements were performed on a treadmill (HP-Cosmos Saturn, Nussdorf-Traunstein, Germany) using a classical racing bicycle (Canondale Advanced CAAD 2, 58 cm, 10.1 kg). The treadmill’s speed and slope were calibrated according to the recommendations of Myers et al. (2009), using a chronometer and a theodolite respectively (Longqiang, LQTS-522D, Xi’an, China). One male participant (age 29 years, height 1.79 m, mass 66.5 kg) was asked to maintain the bicycle in equilibrium on the treadmill without pedalling at a

Correspondence: Y. Henchoz, Institute of Sports Sciences, University of Lausanne, Baˆtiment Vidy, CH-1015 Lausanne, Switzerland. E-mail: [email protected] ISSN 0264-0414 print/ISSN 1466-447X online Ó 2010 Taylor & Francis DOI: 10.1080/02640414.2010.498483

2

Y. Henchoz et al.


constant speed of 5.56 m s71, with a dynamometer fixed in parallel to the surface of the treadmill and holding the bicycle in place in the front. Figure 1 provides a schematic illustration of the procedure and the forces that apply on the bicycle–rider system. For each condition, the method consisted of 11 measurements of the force required to hold the cycle at different treadmill slopes (0–10%, increment 1%). Due to steering inputs required to balance the cycle, some oscillations inevitably occurred and the mean value was noted. The dynamometer (3B Scientific, Hamburg, Germany; range of measurement 0–100 N, accuracy 0.5 N) was calibrated by hanging successively two masses of 5 and 10 kg on it. Ambient temperature was held constant at *258C. Ethical approval for the study was granted by the local ethics committee. Written informed consent was given by the participant. Calculations Treadmill cycling does not involve wind resistance and the resultant opposing force (F), measured with the dynamometer, was equal to the sum of rolling resistance (Rr) and the force due to gravity (Wx): ð1Þ F ¼ Rr þ Wx ¼ Rr þ M g sin tan 1 a ; where M is the sum of the masses of the participant and the bicycle (kg), g is the acceleration due to gravity (9.81 m s72), and a is the treadmill’s slope (%). Rr was the sum of rolling resistances at the rear (Rr1) and front (Rr2) wheels, but was calculated as the coefficient of rolling resistance (Cr) multiplied by the normal force (i.e. Wy or the sum of the normal forces at the rear (Fn1) and front (Fn2) wheels): Rr ¼ Cr Wy ¼ Cr M g cos tan 1 a ð2Þ

Figure 1. Schematic representation of the dynamometer’s (D) measurement of the resultant force (F) opposing rolling resistances at the rear (Rr1) and front (Rr2) wheels and the parallel component (Wx) of the weight of the rider and the bicycle (W). The normal force is distributed between the rear (Fn1) and front (Fn2) wheels and is opposed to the normal component (Wy) of W.

According to equations (1) and (2), the coefficient of rolling resistance could be calculated as: 1 Cr ¼ F M g sin tan 1 a M g cos tan 1 a After reduction, and using the trigonometric property sin a ðcos aÞ1 ¼ tan a: 1 Cr ¼ F M g cos tan 1 a a

ð3Þ

Sensitivity of the method We tested whether the method was sensitive enough to detect an effect of tyre inflation pressure and type of tyre on the coefficient of rolling resistance. A classical foot pump (Specialized Airtool Comp) was used to set tyre inflation pressure at 150, 300, 600, 900 and 1100 kPa. Classical 32-spoked wheels (Mavic XP 21) were equipped with three different types of tyre: Specialized All condition 700 6 28, Michelin Pro 3 Race 700 6 23, and Veloflex Kevlar Open Tubular 700 6 22. Eleven measurements of the resultant opposing force were made for each tyre inflation pressure value and for each type of tyre (165 measurements altogether). Reproducibility of the method To test the reproducibility of the method, a second experimenter repeated the 11 measurements of the resultant opposing force at each tyre inflation pressure value for the 700 6 23 tyres. Statistics For each condition of tyre type and inflation pressure, the mean and standard deviation of the 11 measurements of the coefficient of rolling resistance are presented. A two-way analysis of variance (ANOVA) was used to assess the effect on the coefficient of rolling resistance of tyre type (700 6 28, 700 6 23, 700 6 22), tyre inflation pressure (150, 300, 600, 900, 1100 kPa), and the interaction of these two factors. When a significant effect was found, post hoc comparisons were performed by Tukey tests. Reproducibility was tested by 95% confidence interval (95% CI), as recommended by Nevill and Atkinson (1997). Bias corresponded to the mean of the differences between the values obtained by both experimenters. The limits of agreement were calculated as bias + 1.96 standard deviations of the differences. All analyses were carried out with SPSS v.16.0.1 (Chicago, IL). Statistical significance was set at P 0.05.

Rolling resistance in treadmill cycling Results


Sensitivity of the method The effects of tyre type and tyre inflation pressure on the coefficient of rolling resistance are illustrated in Figure 2. The coefficient of rolling resistance for the 700 6 28 tyres was 0.0114 (0.0007), 0.0079 (0.0009), 0.0068 (0.0005), 0.0063 (0.0006), and 0.0057 (0.0003) for 150, 300, 600, 900, and 1100 kPa respectively. For the 700 6 23 tyres, the coefficient of rolling resistance was 0.0096 (0.0007), 0.0068 (0.0008), 0.0057 (0.0007), 0.0051 (0.0009), and 0.0047 (0.0007) respectively. For the 700 6 22 tyres, the coefficient of rolling resistance was 0.0094 (0.0008), 0.0061 (0.0012), 0.0048 (0.0011), 0.0051 (0.0003), and 0.0049 (0.0006) respectively. ANOVA revealed that the method of measurement of the coefficient of rolling resistance was sensitive enough to detect an effect of both tyre type (F2,150 ¼ 65.860; P 5 0.001) and tyre inflation pressure (F4,150 ¼ 237.640; P 5 0.001). The type 6 inflation pressure interaction almost reached significance (F8,150 ¼ 1.944; P ¼ 0.057). The post-hoc analysis revealed a significant difference between the 700 6 28 tyres and both the 700 6 23 tyres (P 5 0.001) and the 700 6 22 tyres (P 5 0.001), but no significant difference between the 700 6 23 tyres and the 700 6 22 tyres (P ¼ 0.072). The coefficient of rolling resistance for a tyre inflation pressure of 150 kPa was significantly higher than the coefficient of rolling resistance for all other tyre inflation pressures (P 5 0.001). The coefficient of rolling resistance for 300 kPa was significantly higher than that for 600, 900, and 1100 kPa (P 5 0.001). The coefficient of rolling resistance for 600 kPa was significantly higher than that for 1100 kPa (P 5 0.01).

Figure 2. Coefficient of rolling resistance (mean + standard deviation) as a function of tyre type and tyre inflation pressure. *Statistically significant difference compared with 150 kPa (P 5 0.001). #Statistically significant difference compared with 300 kPa (P 5 0.001). {Statistically significant difference compared with 600 kPa (P 5 0.01).

3

Reproducibility of the method The measurement of the coefficient of rolling resistance for the 700 6 23 tyres by two distinct experimenters resulted in a small bias of 0.00029 (95% CI, 70.00011 to 0.00068). Values were not log transformed due to the homoscedastic nature of the data. Discussion The present results show that this new method had excellent sensitivity and reproducibility as well as being technically simple and affordable. It is an attractive way of measuring rolling resistance in treadmill cycling. It is, however, not applicable in field cycling but other methods (cf. Introduction), which take into account the compactness of the ground, can be applied. When calculating mechanical power output according to the method described by Coleman et al. (2007), the extrapolation of the coefficient of rolling resistance instead of a precise measurement results in an imprecision, especially at lower speeds, which can be alleviated by using the present method. The amount of the imprecision can be better assessed by taking a numerical example. For a rider of 75 kg cycling on a 1% slope treadmill with a 10 kg bicycle at a speed of 10 m s71, the total force to be overcome is given by equation (1). For a coefficient of rolling resistance of 0.003, the corresponding total force is 10.8 N and the corresponding power output is 108.4 W. For a coefficient of rolling resistance of 0.004, the corresponding total force would be 11.7 N and the corresponding power output would be 116.8 W, giving an imprecision of 7.7%. For a coefficient of rolling resistance of 0.005, the imprecision would be 15.4%. The shape of the relationship obtained between the coefficient of rolling resistance and tyre inflation pressure in Figure 2 is comparable to that obtained previously (Grappe et al., 1999), thereby confirming that the coefficient of rolling resistance is inversely and hyperbolically related to tyre inflation pressure. Few studies have reported of coefficients of rolling resistance on a treadmill: Davies (1980) found a very low value of 0.001, without mentioning tyre type and inflation pressure. Me´nard (1992) obtained values of 0.0028–0.0058 for different tubular tyres with inflation pressures of 300–500 kPa. From a practical point of view, several sources of error are likely to lower the accuracy of this method. First, the treadmill’s slope was varied from 0% to 10% and it was expected that the change in the treadmill’s slope between the repetitions of a single condition would affect only the force due to gravity, and that rolling resistance would remain constant. It


4

Y. Henchoz et al.

is questionable whether the decrease in the normal force due to the increase in the slope had an impact on the coefficient of rolling resistance. Indeed, Grappe et al. (1999) found a second-order polynomial relationship between the coefficient of rolling resistance and an increase in the normal force. The coefficient of rolling resistance increased from 0.0035 to 0.0040 with an overload of 15 kg on a participant of 66.2 kg and a bicycle of 9.8 kg (Grappe et al., 1999). In the present study, according to equation (2), an increase of the slope from 0% to 10% caused a change of the normal force from 751.45 N to 747.72 N for a participant of 66.5 kg on a bicycle of 10.1 kg. According to the relationship between the coefficient of rolling resistance and the normal force obtained by Grappe et al. (1999), this minor change corresponds to a coefficient of rolling resistance change of only 6.86 6 1077, which is negligible. Second, the dynamometer used was accurate to 0.5 N. Implementing this value into equation (3) would lead to an error of 0.0007 in the coefficient of rolling resistance. Third, another source of error might arise from the treadmill slope, which was indicated with 0.1% accuracy. A difference of 0.05% would induce an error of 0.0005 in the coefficient of rolling resistance. However, it should be specified that 0.0007 and 0.0005 are maximal possible errors. Due to the 11 measurements at different slopes for each condition, the positive errors were partly offset by the negative errors and the mean coefficients of rolling resistance were obtained with more accuracy than each measurement individually. To conclude, the present method is sensitive, reproducible, and provides an attractive way of measuring rolling resistance in treadmill cycling. In addition, future investigations on the effects of different parameters such as speed, additional mass

and tyre characteristics on the coefficient of rolling resistance are possible with a well-controlled treadmill set-up. References Candau, R. B., Grappe, F., Menard, M., Barbier, B., Millet, G. Y., Hoffman, M. D. et al. (1999). Simplified deceleration method for assessment of resistive forces in cycling. Medicine and Science in Sports and Exercise, 31, 1441–1447. Capelli, C., Rosa, G., Butti, F., Ferretti, G., Veicsteinas, A., & di Prampero, P. E. (1993). Energy cost and efficiency of riding aerodynamic bicycles. European Journal of Applied Physiology and Occupational Physiology, 67, 144–149. Coleman, D. A., Wiles, J. D., Davison, R. C., Smith, M. F., & Swaine, I. L. (2007). Power output measurement during treadmill cycling. International Journal of Sports Medicine, 28, 525–530. Davies, C. T. (1980). Effect of air resistance on the metabolic cost and performance of cycling. European Journal of Applied Physiology and Occupational Physiology, 45, 245–254. di Prampero, P. E., Cortili, G., Mognoni, P., & Saibene, F. (1979). Equation of motion of a cyclist. Journal of Applied Physiology, 47, 201–206. Grappe, F., Candau, R., Barbier, B., Hoffman, D., Belli, A., & Rouillon, J. D. (1999). Influence of tyre pressure and vertical load on coefficient of rolling resistance and simulated cycling performance. Ergonomics, 10, 1361–1371. Me´nard, M. (1992). L’ae´rodynamisme et le cyclisme. In Jornadas Internacionales sobre Biomecanica del Ciclismo, Tour 92. Donostia San Sebastian: Centro de estudios e investigaciones tecnicas de gipuzkoa. Myers, J., Arena, R., Franklin, B., Pina, I., Kraus, W. E., McInnis, K. et al. (2009). Recommendations for clinical exercise laboratories: A scientific statement from the American Heart Association. Circulation, 119, 3144–3161. Nevill, A. M., & Atkinson, G. (1997). Assessing agreement between measurements recorded on a ratio scale in sports medicine and sports science. British Journal of Sports Medicine, 31, 314–318. Pugh, L. G. (1974). The relation of oxygen intake and speed in competition cycling and comparative observations on the bicycle ergometer. Journal of Physiology, 241, 795–808. Whitt, F. R. (1971). A note on the estimation of the energy expenditure of sporting cyclists. Ergonomics, 14, 419–424.