European Journal of Applied Physiology © Springer-Verlag 2008 10.1007/s00421-008-0692-z Original Article [/url]Effects of short-term training using SmartCranks on cycle work distribution and power output during cycling
Harald Böhm1
, Stefan Siebert1 and Mark Walsh2 [/url](1) Department of Sports Equipment and Materials, Faculty of Sport Science, University of Technology Munich, Connollystr 32, 80809 Munich, Germany [/url](2) Physical Education, Health and Sports Studies Department, Miami University, Oxford, OH 45056, USA
[/url]
Harald Böhm Email: boehm@tum.de Accepted: 30 January 2008
Published online: 14 February 2008 [/url]Abstract SmartCranks use a free running bearing to promote independent pedal work by each leg during cycling. This system is designed for training the upstroke phase during cycling. The effects of training with SmartCranks on the power output (PO) and on cycle work distribution at the anaerobic threshold and the maximum power level were examined. Twenty male, non-professional cyclists were randomly assigned into intervention and control group, training 5 weeks with SmartCranks and conventional cranks, respectively. Before and after the training period the subjects performed an incremental test to exhaustion. Lactate was measured to determine the individual anaerobic threshold (IAT) and forces at the pedal were recorded to quantify changes in the work distribution over the full revolution. We observed no significant statistical difference for peak power (PO; 333.3 ± 32.8 W vs. 323.3 ± 21.8 W) and PO at IAT (229.6 ± 30.1 W vs. 222.7 ± 25.2 W) for SmartCrank and control conditions, respectively (
P > 0.05). However, we did observe that work distribution in the downward phase was significantly reduced in the SmartCranks training group at peak PO (from 70.0 ± 4.9% to 64.3 ± 5.8%;
P < 0.05). Although the possible implications of the change in the work distribution of sectors are not known, for the success in cycling performance—indicated by the PO—training with the SmartCranks was not more advantageous than training with conventional bicycle cranks.
Keywords Upstroke phase - Work load - Pedal-crank system - Power output [/url]
Introduction
To improve the transfer of human power into cycling performance, technical solutions such as elliptic chain rings, pedal-crank systems with varying lengths or independent crank arms (rotor system) have been developed. Since the effects of these systems are small (Hull et al.
1992; Hue et al.
2001; Zamparo et al.
1981; Santalla et al.
2002; Lucia et al.
2004), the present mechanism with fixed crank length and circular chain rings are most commonly used in cycling. In the absence of better equipment, it is possible for the athlete to improve pedaling technique. The task of the athlete is hereby to maximize the motive efficiency, defined as the ratio between propulsive tangential force and the total force applied to the crank (Davis and Hull
1981). It has been shown that applying optimal oriented forces to the pedal during cycling enhances power output (PO) for comparable load magnitudes (Davis and Hull
1981). Furthermore, motive efficiency plays a determinant role in modulating muscular efficiency changes at a constant pedaling rate (Zameziati et al.
2006). To help the athletes in improving their pedaling technique, visual feedback of the tangential pedaling force has been used (Sanderson
1987; Sanderson and Cavanagh
1990). Another method to improve motive efficiency is to use SmartCranks (SmartCranks GmbH, Zug, Switzerland, Fig.
1), also known under the name PowerCranks (PowerCranks Inc., CA, USA). For the sake of clarity in this manuscript, we will use the term “SmartCranks” to represent literature reporting on either the SmartCrank or PowerCrank. SmartCranks use free-running bearings in each crank arm to promote independent pedal work by each leg during cycling. Each leg can drive the bicycle but one leg cannot assist the other. Particularly in the upward phase where motive efficiency is low or even negative (Davis and Hull
1981; Zameziati et al.
2006), training with SmartCranks might improve motive efficiency for two reasons: first, intensified training of the hip flexors and hamstring muscles allows these muscles to contribute to the propulsive tangential force produced in the upward phase. Second, movement variations introduced by the SmartCranks might cause a greater sensitivity of the neuromotor system (Davids et al.
2003; Hodges and Franks
2002) to produce maximization of the motive efficiency. [/url]
Fig. 1 SmartCranks training system
Luttrell and Potteiger (
2003) showed a significant increase in gross efficiency (GE) after training with SmartCranks compared to conventional crank system; however, Luttrell and Potteiger did not investigate if the increase in GE is related to a change in motive efficiency or to the physiological changes that occurred during training by the greater muscle involvement of the hip and knee flexors. Motive efficiency assumes by its definition that radial forces (Fig.
2), which do not perform any mechanical work during the cycle, should be avoided. To our best knowledge it has not been shown that the radial forces are of disadvantage to the performance of the athlete. Radial forces might be a necessary compromise of the musculoskeletal system for generating considerable tangential forces, due to indirect transfer of muscular action via resultant joint torques (Böhm et al.
2006). In cycling, an active pull in the upward phase leads to greater motive efficiency by increasing the tangential force in the upward phase. This enhancement of motive efficiency can be described by the work produced in the upward phase without considering radial forces. For that reason work distribution is an alternative well-defined physical parameter to estimate motive efficiency. [/url]
Fig. 2 SRM System to control the ergometer resistance and force measurement system mounted to the crank. Two components, the radial and the tangential forces to the crank can be resolved
The aim of this study was therefore to investigate whether training with the SmartCranks system has a significant effect on first, the PO at the anaerobic threshold and the maximum power level, and second, on the work distribution during revolution. [/url]
Methods [/url]Subjects and training
A group of 20 male, healthy, non-professional cyclists (28 ± 7 year, 76 ± 8 kg, 180 ± 5 cm) volunteered to participate in this study. All subjects had practice in cycling with click pedals. Because previous studies showed that the grade of physiological fitness has no effect on motive efficiency (Sanderson
1991), no emphasis was placed on the proficiency level of the subjects participating in the study. Subjects were randomly assigned into intervention and control groups. The intervention group trained with the SmartCranks system and the controls with a conventional crank system. Crank length (175 mm) and chain rings (53 teeth) were similar on the SmartCranks and the conventional cranks system.
The training program took 5 weeks with two training sessions per week, which were executed on racing bicycles on indoor cycling trainers. The training was designed heeding the common precepts for motor skill training (Schmidt and Wrisberg
1991; Bompa
1999). The weekly training was divided into two sessions. In the first session, eight repetitions with duration of 3 min at 50 rpm and 80% of individual anaerobic threshold (IAT) and total recovery (at 90 rpm and 40% IAT) between the repetitions were performed. In the second session ten repetitions with a 3-min duration of at varying pedaling rates (from 60 rpm up to the individual max.) and 70% of IAT and total recovery between the repetitions was performed. During recovery, warm up and cool down, the intervention group had to lock their SmartCranks free-running bearing. There was a 48-h recovery break between the two weekly training sessions. Every week the number of repetitions was decreased while the duration of each repetition was increased. Additionally the total volume of the training sessions was increased from 60 to 90 min from the beginning to the end of the training period. Intensity of training was individually set by the heart rate of the athletes according to their individual pre-test results. A considerably high intensity (resistance of the cycling trainers) was chosen (70–80% IAT level) to obtain a better motor feedback (Bompa
1999). [/url]Measurements and instrumentation
Before and after the training period, the subjects performed an incremental test to exhaustion, starting at 100 W PO level, with 30 W increments every 3 min, at a constant pedaling rate of 90 rpm. A cycle ergometer (SRM, Schoberer GmbH, Germany) with an eddy current brake was used. The ergometer was equipped with racing-style handlebar and saddle, which were adjusted to fit the subjects’ individual body geometry. Saddle height was adjusted by sitting on the saddle in the normal riding style. When resting the heel on the pedal with the crank arm in the lower dead point (180°), the knee angle should be about 5°. The horizontal position of the saddle was adjusted in the way that the subject’s patella was perpendicular to the axle of the pedal, with the crank arm in 90° position. Horizontal and vertical position of the handlebar was adjusted according to the subject’s individual racing bicycle. Subjects used the same pedal and shoe system for both testing and training. Heart rates were continuously monitored during the test (Polar S625X, Polar Electro OY, Finland). In the last 15 s of each step 20 μl blood was taken from the subject’s earlobe to determine blood lactate concentration with an enzymatic lactate-analyzer (Biosen 5040, EKF GmbH, Germany). IAT was determined at the 45° tangent to the lactate curve with increasing power levels according to Simon et al. (
1981).
Tangential and radial forces were recorded over 150 Cycles on each incremental step, starting 30 s after the beginning of the step and ending before lactate and heart rate were taken. Tangential and radial forces applied to the pedal were sampled at 1,000 Hz with crank adapters (Powertec, O-tec GmbH, Germany) mounted to the SRM system (Fig.
2). The Hall sensors in the adapters were measuring the deformation of the crank caused by the force applied to the pedal. The crank angle was recorded by incremental signals occurring 12 times per revolution caused by magnetic transducers each separated at 30°. [/url]Calculation of work distribution and pedaling effectiveness Work distribution was calculated over four sectors representing downward, backward, upward and forward directions of motion (Fig.
3). Calculation of the work during the downward sector is shown in Eq. 1. [/url]
(1) [/url]
Fig. 3 Right pedal directional action of the cyclist is subdivided into four sectors:
forward,
downward,
backward and
upward
The crank length (
l crank) was fixed to 175 mm during the whole test. To represent the work distribution among different sectors in a concise way, the work in each sector was normalized to the total work produced during the full revolution. To compare the results to previous investigations, the index of effectiveness (IE) for the different sectors was calculated according to Coyle et al.
1991 with Eq. 2 shown for the downward sector. [/url]
(2)
Tangential and radial forces
F tangenital and
F radial of each subject were averaged over about 150 revolutions on each incremental step before they were used to calculate sectors work and IE. [/url]Statistical analysis
The athlete’s IAT and maximum PO as well as the normalized work distribution at both power levels underwent statistical testing. The Kolmogorov–Smirnov test was applied to ensure a Gaussian distribution of the results. A group by time analysis of variance (ANOVA) with repeated measures on time (pre-, post-training) was performed to analyze IAT and maximum PO. A group by time and power level ANOVA with repeated measures on time (pre-, post-training) and power level (IAT, maximum) was performed to analyze sector work distribution. Statistical significance was set at the 0.05 level for all tests. [/url]
Results [/url]Power output Mean values corresponding to all POs in the pre- and post-tests for the different training groups are shown in Fig.
4 and Table
1 for the IAT and in Fig.
5 and Table
2 for the maximum PO. [/url]
Fig. 4 Mean power output at IAT power level for the SmartCranks and control group in the pre- and post-tests
[/url]Table 1 Mean values of power output (PO) and sectors work distribution for the IAT power level
Parameter
SmartCranks group
Control group
Pre
Post
Pre
Post
Fore (%)
22.1 ± 3.5
17.8 ± 2.1
21.5 ± 2.0
17.7 ± 2.5
Down (%)
83.6 ± 4.4
79.9 ± 7.9
91.9 ± 10.6
83.4 ± 5.7
Back (%)
13.6 ± 3.3
20.1 ± 2.9
11.1 ± 4.6
20.5 ± 4.4
Up (%)
−19.4 ± 6.2
−17.8 ± 6.8
−24.6 ± 8.2
−21.6 ± 5.8
PO (W)
218.7 ± 36.3
229.6 ± 30.1
213.3 ± 31.9
222.7 ± 25.2
IAT (%max)
68.8 ± 5.7
68.7 ± 3.6
67.92 ± 6.0
68.8 ± 4.8 [/url]
Fig. 5 Mean power output at maximum power level for the SmartCranks and control group in the pre- and post-tests
[/url]Table 2 Mean values of power output (PO) and sectors work distribution for the maximum power level
Parameter
SmartCranks group
Control group
Pre
Post
Pre
Post
Fore (%)
21.7 ± 2.5
19.3 ± 2.6
21.3 ± 1.8
17.7 ± 2.8
Down (%)
70.0 ± 4.9
64.3 ± 5.8
72.5 ± 9.0
69.5 ± 3.3
Back (%)
15.8 ± 3.1
20.6 ± 2.6
14.0 ± 5.7
19.6 ± 2.4
Up (%)
−7.5 ± 4.1
−4.1 ± 4.2
−7.8 ± 3.6
−6.9 ± 3.9
PO (W)
320.0 ± 30.0
333.3 ± 32.8
313.3 ± 31.6
323.3 ± 21.8
For the PO at IAT level, the two factor ANOVA showed a significant time (training) effect [
F(1,16) = 13.22,
P = 0.002]. The partial eta-squared value (
η 2) for this significant effect was 0.45 indicating a moderate effect size (Cohen
1988). No significant interaction effect (time × crank system used during training) was found [
F(1,16) = 0.08,
P = 0.783,
η 2 = 0.005].
For the maximum PO a significant time effect [
F(1,16) = 7.00,
P = 0.018] with a moderate effect size
η 2 = 0.30 was observed. No significant interaction effect (time × crank system used during training) was found [
F(1,16) = 0.143,
P = 0.710,
η 2 = 0.009]. [/url]Work distribution of sectors Calculations of the work distribution (Eq. 1) were based on the tangential forces measured over the full revolution shown in Fig.
6. Considering the first seven incremental steps (100–280 W), all subjects’ average maximum tangential forces occur at 94°. The maximum was located within the downward sector, which had the highest work contribution of 43–72 J shown in Fig.
7. The minimum tangential force occurred at a plateau between 260° and 280° and pointed in the opposite direction of the revolution and contributed to a negative work of −22 to −11 J in the upward sector. Normalizing the sectors’ work distribution to the total work, the forward and backward sectors’ contribution remained almost constant at 24 and 12%, respectively, whereas the work contribution of the downward sector decreases from 129 to 77%, while at the same time those of the upward sector increases from −65 to −12%. [/url]
Fig. 6 Average tangential and radial forces measured on all 20 Subjects for the
right crank. Shown are the seven steps ranging from 100 to 280 W, which all subjects could reach during the incremental pre-test performed
[/url]
Fig. 7 Average work of all 20 subjects, performed during the first seven steps of the incremental pre-test over four sectors during the cycle of the right crank. The
lower graph shows the work normalized over the full revolution
Mean values corresponding to all the sectors’ work distributions in the pre- and post-tests of the SmartCranks and the control group are shown in Fig.
8 and Table
1 for the IAT and in Fig.
9 and Table
2 for the maximum power level. Training reduced the work performed in the forward and downward sectors and increased the work performed in the backward and upward sectors at both power levels and for all training groups. This was confirmed statistically because the three-factor ANOVA showed a significant time effect for all four sectors: the forward [
F(1,16) = 71.70,
P < 0.001,
η 2 = 0.81], the downward [
F(1,16) = 11.75,
P = 0.003,
η 2 = 0.42], the backward [
F(1,16) = 32.95,
P < 0.001,
η 2 = 0.67] and the upward sectors [
F(1,16) = 8.64,
P = 0.010,
η 2 = 0.35]. The
η 2-values showed a moderate to strong effect size. [/url]
Fig. 8 Sectors work distribution at the IAT power level for the SmartCranks group and the controls at the pre- and post-test
[/url]
Fig. 9 Sectors work distribution at the maximum power level for the SmartCranks group and the controls at the pre- and post-test
The power level had a significant effect for the downward [
F(1,16) = 194.20,
P < 0.001,
η 2 = 0.92], backward [
F(1,16) = 9.58,
P = 0.007,
η 2 = 0.38] and upward sectors [
F(1,16) = 209.91,
P < 0.001,
η 2 = 0.93], but no significant effect was observed for the forward sector [
F(1,16) = 0.27,
P = 0.611,
η 2 = 0.02].
A significant interaction effect (time × power level) was only observed in the backward sector [
F(1,16) = 9.13,
P = 0.008,
η 2 = 0.36]. No significant interaction effect (time × crank system used during training, or level × crank system used during training) for any sector was observed.
The interaction effect (time × power level × crank system used during training) was significant for the downward sector [
F(1,16) = 4.59,
P = 0.048] with a moderate effect size (
η 2 = 0.22). The other sectors showed no significant interaction effects of all three factors. [/url]
Discussion
The results of this study demonstrate that the crank system used during the 5-week training had no significant effect on the PO at IAT and maximum level reached in the incremental test to exhaustion. Since PO at IAT and maximum power level are considered as strong predictors of success in endurance events and sprint cycling, respectively (Coyle et al
1992; Kumagai et al.
1982), this study showed that training with the SmartCranks was not advantageous than training with conventional bicycle cranks.
To calculate the work distribution during revolution tangential forces were measured at the pedal during the incremental test to exhaustion. The forces measured in this study were different in magnitude to those measured by Zameziati et al. (
2006) cycling at a comparable power level of 160 W with a slightly lower cadence of 80 rpm, instead of 90 rpm. The author’s average maximum force of 270 N occurred at 89°. In this study the average maximum force of 244 N was 26 N lower and was located at 94° (Fig.
6). The average minimum tangential force measured by Zameziati was −50 N; it occurred around 240°. In this study the minimum force as well as the maximum force was about 25 N lower. The difference might have been a result of the higher pedaling frequencyand the measured population or measurement errors. To eliminate any possibility of calibration errors, the total work of the tangential force was calculated and the sum over the four sectors for the right crank was 53 J (Fig.
7). At 90 rpm this corresponds to 79.6 W whose double amount comes up to 159.2 W and corresponds well to the power setting at the ergometer. Due to this test we were confident about the forces measured and consequently about the work calculated from the tangential forces in the different sectors.
The statistical analysis revealed that both factors, training as well as the power level had a significant effect on the work distribution for both groups. A significant synergistic effect of the training with the SmartCranks system was observed only in the downward sector at the maximum power level. Compared to the control group, the work produced by the SmartCranks group was lower in the downward phase (Fig.
9). To obtain the same effect on PO as the control group, the work of the downward sector must be distributed to the other three sectors. This result corresponds to those expected from the SmartCrank mechanism, built for especially training of the hip and knee flexor muscles used predominantly in the upward sector. This might explain the results of Luttrell and Potteiger (
2003), which showed a significant increase in GE after training with SmartCranks compared to conventional crank system. The changes in GE observed by Luttrell and Potteiger during training could be caused by the greater muscle involvement of the hip and knee flexor muscles during the recovery phase. Their increase in strength may contribute to the increase in GE values observed, with less oxygen needed at the same workload. The question arises, however, about the effectiveness of pulling up on the cranks during the recovery phase when cycling, and whether this offers any advantage to the cyclist. In this study as well as in the study of Luttrell and Potteiger (
2003) PO at IAT level was not increased when training with SmartCranks. The reason for this behavior might be the significant decrease of the work produced in the downward phase, which seems to be a drawback of the SmartCranks system. The muscles primarily involved during the downward phase are the muscles of the quadriceps group, the gluteus maximus, and the gastrocnemius (Faria
1992). These knee extensor muscles are the strongest muscles of the lower extremity (Böhm et al.
2006), and their contribution to the downward phase was within a most effective tangential direction, which was demonstrated by the index of efficiency of about 90% in the downward sector (Fig.
10). Since the knee extensors are well suited for cycling, the shift of the work from the knee extensors during the downward phase to a greater muscular action in the recovery phase is questionable. The reason for the reduced work in the downward sector for the SmartCranks group might be increased co contraction caused by increased muscular action in the transition between upward and downward phases. Another reason might be the missing recovery phase for gastrocnemius, a two-joint muscle, acting both as knee flexor in the upward phase and as ankle flexor in the downward phase. [/url]
Fig. 10 Average work distributions and index of effectiveness of 20 subjects during incremental pre-test to exhaustion. To compare the different numbers of power levels reached by the subjects, the power is first normalized on the maximum power reached by each athlete. Nine equally spaced power steps ranging from 34 to 100% of maximum individual power are then calculated by interpolating in between the increasing power levels measured during the test
In this study a new approach of work distribution among sectors during the cycle instead of the commonly used IE was performed. The difference between both calculation methods can be demonstrated clearly when looking at the change of the parameters in the four sections with increasing power levels (Fig.
10). The changes in work distribution occurred mainly in the downward and upward sectors; IE increased in all sectors except the downward sector where it remains constant at about 90%. Unlike the IE, the work distribution in the downward phase was decreasing from 120 to 72% about a considerable amount of 48% of the work performed during full revolution. In previous studies on the relation of motive efficiency and muscular efficiency little attention was paid on the down stroke phase, since the calculated IE showed no variation by the interventions performed (Coyle et al.
1991; Zameziati et al.
2006). The IE in the downward phase represents only the effectiveness of the force applied in tangential direction to the crank and not its work contribution to the revolution which is important for the PO and therefore, for the efficiency of the muscular system. For future studies dealing with motive efficiency we suggest in addition to IE taking into account the work distribution. Work distribution gives insight in the total amount of work performed in the sectors, which especially in the downward sector—considering its excellent force effectiveness of about 90%—is valuable information when it is reduced by training as it was observed in this study.
In conclusion, this study demonstrates that the crank system used during the 5-week training had no significant effect on the PO at IAT and maximum level. Furthermore, the results showed that the SmartCranks training group produced significantly less work in the downward sector compared to the controls. This was compensated by the other sectors to obtain the same effect on PO as the control group, but is highly questionable whether a decrease in the downward sector—considering its high force effectiveness—is a benefit for the cyclist. Acknowledgments We thank the SmartCranks GmbH and the TOUR-magazine for providing the equipment used in the study.
[/url]References [/url]Böhm H, Cole GK, Brüggemann GP, Ruder H (2006) Contribution of muscle series elasticity to maximum performance in drop jumping. J Appl Biomech 22:3–13
[/url]Bompa T (1999) Periodization, theory and methodology of training. Human Kinetics, Champaign, pp 60–63 [/url]Cohen J (1988) Statistical power analysis for the behavioral sciences, 2nd edn. Lawrence Erlbaum, Hillsdale, pp 20–28 [/url]Coyle EF, Feltner ME, Kautz SA, Hamilton MT, Montain SJ, Baylor AM, Abraham LD, Petrek GW (1991) Physiological and biomechanical factors associated with elite endurance cycling performance. Med Sci Sports Exerc 23:93–107
[/url]Coyle EF, Sidossis SF, Horowitz JF, Beltz JD (1992) Cycling efficiency is related to the percentage of Type I muscle fibers. Med Sci Sports Exerc 24:782–788
[/url]Davids K,Glazier P, Araujo D, Bartlett R (2003) Movement systems as dynamical systems. The functional role of variability and its implications for sports medicine. Sports Med 33:245–260
[/url]Davis RR, Hull ML (1981) Measurement of pedal loading in bicycling: II. Analysis and results. J Biomech 14:857–872
[/url]Faria IE (1992) Energy expenditure, aerodynamics and medical problems in cycling: an update. Sports Med 14:43–63
[/url]Hodges NJ, Franks IM (2002) Modelling coaching practice: the role of instruction and demonstration. J Sports Sci 20:793–811
[/url]Hue O, Galy O, Hertogh C, Casties JF, Prefaut C (2001) Enhancing cycling performance using an eccentric chainring. Med Sci Sports Excerc 33:1006–1010
[/url]Hull ML, Williams M, Williams K, Kautz S (1992) Physiological response to cycling with both circular and noncircular chainrings. Med Sci Sports Exerc 24:1114–1122
[/url]Kumagai S, Tanaka K, Matsuura Y, Matzusaka A, Hirakoba K, Asano K (1982) Relationships of the anaerobic threshold with the 5 km, 10 km, and 10 mile races. Eur J Appl Physiol 49:13–23
[/url]Lucia A, Balmer J, Davison RCR, Perez M, Santalla A, Smith PM (2004) Effects of the rotor pedalling system on the performance of trained cyclists during incremental and constant-load cycle-ergometer tests. Int J Sports Med 25(7):479–485
[/url]Luttrell MD, Potteiger JA (2003) Effects of short-term training using PowerCranks on cardiovascular fitness and cycling efficiency. J Strength Conditioning Res 17:785–791
[/url]Santalla A, Manzano JM, Perez M, Lucia A (2002) A new pedaling design: the rotor-effects on cycling performance. Med Sci Sports Exerc 34:1854–1858
[/url]Sanderson DJ (1987) Training with biofeedback. Real-time feedback improves pedalling style. Bike Tech 6:10–13 [/url]Sanderson DJ (1991) The influence of cadence and power output on the biomechanics of force application during steady-rate cycling in competitive and recreational cyclists. J Sports Sci 9(2):191–203
[/url]Sanderson DJ, Cavanagh PR (1990) Use of augmented feedback for the modification of the pedaling mechanics of cyclists. Can J Sport Sci 15:38–42
[/url]Schmidt R, Wrisberg C (1991) Motor learning and performance. Human Kinetics, Champaign [/url]Simon G, Berg A, Dickhuth H, Simon-Alt A, Keul J (1981) Determination of anaerobic threshold depending on age and performance capacity. Dtsch Z Sportmed 32:7–14 [/url]Zameziati K, Mornieux G, Rouffet D, Belli A (2006) Relationship between the increase of effectiveness indexes and the increase of muscular efficiency with cycling power. Eur J Appl Physiol 96:274–281
[/url]Zamparo P, Minetti A, Prampero P (1981) Mechanical efficiency of cycling with a new developed pedal-crank. J Biomech 35:1387–1398
, Stefan Siebert1 and Mark Walsh2 [/url](1) Department of Sports Equipment and Materials, Faculty of Sport Science, University of Technology Munich, Connollystr 32, 80809 Munich, Germany [/url](2) Physical Education, Health and Sports Studies Department, Miami University, Oxford, OH 45056, USA