Journal Logo

APPLIED SCIENCES: Physical Fitness and Performance

Testing peak cycling performance: effects of braking force during growth


Author Information
Medicine & Science in Sports & Exercise: February 2000 - Volume 32 - Issue 2 - p 493
  • Free


The optimal braking force (resistance) should elicit the highest value possible for cycling anaerobic power. Identification of an appropriate braking force (FB) is difficult during growth and maturation because of the changes in body composition. Few studies have investigated the optimal resistance required for children. Dotan and Bar-Or (6), studying the Wingate protocol, reported FB values for active 13- to 14-yr-old girls and boys of 0.66 N·kg1 and 0.69 N·kg1 body mass (BM), respectively. In the same study, optimal FB in adults was calculated as 0.83 N·kg1 BM in women and 0.86 N·kg1 BM in men. Carlson and Naughton (3) proposed FB values between 0.60 and 0.80 N·kg1 BM in 6- to 12-yr-old girls and boys. Van Praagh et al. (24,25) determined the optimal resistance to elicit cycling peak power (CPP) for individual subjects from the force-velocity test. The need to adjust FB for young people at different stages of maturation is highlighted by the finding of a mean optimal resistance of 0.63 N·kg1 BM in 7-yr-old boys (24) and 0.67 and 0.83 N·kg1 BM in 12- to 13-yr-old girls and boys, respectively (25). However, considering that the above-mentioned studies did not incorporate the force required to accelerate the flywheel of the cycling device (inertial force) in their respective calculations, the maximal power reported was undoubtedly underestimated (13). In contrast to previous protocols, the imposed total external force (defined as FB + inertial force) is not constant during the acceleration of the flywheel. Therefore, recent concomitant measurements of forces, velocities, and powers were reported during the acceleration phase of sprint cycling. These studies showed that maximal cycling power (flywheel inertia included) measured during an all-out sprint or calculated from the power-velocity relationship was independent of the FB applied in young adults (1,14,21) as well as in adolescents (5). This new test protocol allows the measurement of maximal power and optimal velocity from only one sprint and does not require high resistances to be overcome. Recently, Martin et al. (16) measured CPP in one short exercise bout with resistance provided solely by the moment of inertia of the flywheel. Moreover, CPP (flywheel inertia included) is obtained during the acceleration phase of the sprint and, thus, much earlier than in the “classical” force-velocity cycling test. This optimal procedure of one short-term sprint limits motivational factors and muscle fatigue, which are responsible for the rapid decrease of CPP (18), and is also less time consuming. To our knowledge, no studies using the present method have investigated CPP and its relationship with the applied FB in prepubescent children. To elucidate this issue, the purpose of the present study was to investigate the relationship between peak power (PP; flywheel inertia included) and FB in nonathletic male children, adolescents, and adults.


Subjects and material.

A total of 520 male subjects aged 8–20 yr volunteered for the study with parents’ written informed consent. BM and standing height were assessed. Lean leg volume (LLV) was estimated by anthropometry (11).

The exercise bouts were conducted on a calibrated friction-loaded ergometer (Ergomeca, Sorem, Toulon, France) that had the following features: 0.17-m crank length, 6.12-m excursion at the perimeter of the flywheel per pedal revolution, an arm balance allowing frictional forces between 0 and 196.2 N (i.e., braking loads between 0 and 20 kg), 15.6-kg flywheel mass, and 0.26-m flywheel radius. Two optical sensors were mounted on the ergometer to detect the onset of the crank gear rotation cycle and the measurement of the rolling speed of the flywheel (8). Values of instantaneous velocity, force, and power were calculated four times per flywheel revolution (16 counts per pedal revolution) and were averaged per half a pedal revolution.

Testing procedure.

The warm-up consisted of 2-min submaximal cycling against a low FB followed by a brief sprint (5 s) at the same FB as in the first exercise bout. Each subject had to perform three “all-out” sprints, each separated by a period of at least 4-min rest. The subject was told to remain in the saddle and was vigorously encouraged to produce the highest acceleration. Toe clips were used to prevent the feet from slipping.

Three FB were applied in a random order according to the individual BM: 0.245, 0.491, and 0.736 N·kg1 BM (corresponding applied loads: 25 [FB25], 50 [FB50], and 75 [FB75] g·kg1 BM, respectively).


The total external force exerted by the subject was equal to the sum of the frictional force (Ffrict) against the FB applied to the flywheel, and the inertial force (Finert) required to accelerate the latter (13).

The instantaneous power output (Pinst, in W) generated during the sprint was computed as follows:


It was therefore necessary to determine the flywheel inertia of the device. According to Lakomy (13) the flywheel inertia was calculated from the linear relationship between the free deceleration of the flywheel and the friction force applied. For the present cycle ergometer, the relationship between force (N) and acceleration (rpm·s1) was as follows:


This equation indicates a moment of inertia of 0.546 kg·m2 for the flywheel. Our results were very close to those obtained by Linossier et al. (14) for the same type of cycle ergometer (0.517 kg·m2).

To quantify one complete push-off phase that corresponds to the contraction cycle of each muscle group, velocity, force, and power were averaged per half-pedal revolution. PP (PP25, PP50, and PP75), corresponding optimal velocity (v25, v50, and v75), and force (F25, F50, and F75) were determined for each selected FB (FB25, FB50, and FB75). For each individual, the highest PP was considered as CPP. Optimal velocity (vopt) corresponded to the velocity at CPP (Fig. 1). Measurements also included the time to reach PP (tPP25, tPP50, and tPP75). The difference between each PP and CPP was expressed as follows:

Figure 1
Figure 1:
An example of power-velocity relationships for a typical child. FB, braking force; CPP, cycling peak power; PP75, peak power at FB 0.736 N·kg\-1 BM; v75, velocity at FB 0.736 N·kg\-1 BM; vopt, optimal velocity.


Statistical analysis.

Polynomial regressions were used to determine the relationships between anthropometric variables and age for all subjects. The subjects were then divided into age groups (Table 1). Differences between age groups for performance variables (maximal power, vopt, optimal force, and time to reach CPP) were tested by ANOVA. A contrast test (protected list of significant difference of Fisher) was performed when the ANOVA F-ratio was significant (P < 0.05). The difference between PP for each FB (PP25, PP50, and PP75) and CPP (DPP) was studied. The relationships between DPP and age were fitted using a third-order polynomial regression. The same statistical analysis was used to study the evolution of Dv [(v –vopt)·vopt–1 %], DF [(FFoptFopt–1 %], and DtPP [(tPP –tCPP)·tCPP–1 %] with growth.

Table 1
Table 1:
Subjects in different age groups (mean ± SEM).


BM and LLV increased with age. However, the results plateaued between 8 and 10 yr and after 17 yr (Fig. 2). Relationships between both anthropometric variables and age were described by the following third-order polynomial regressions:

Figure 2
Figure 2:
Relationship between body mass (BM), lean leg volume (LLV) characteristics, and age. ***P < 0.001


The LLV/BM ratio increased with age (N = 531, r = 0.61, P < 0.0001) from 0.075 ± 0.001 for children aged 9 yr to 0.095 ± 0.001 for young adults aged 19 yr.

The same profile as for the anthropometric variables was observed for absolute (W) and relative CPP (W·kg1 BM):MATH 6

vopt also increased until 17 yr, while the time to reach CPP (tCPP) decreased during growth (P < 0.0001) until 15 yr and plateaued thereafter (Fig. 3).

Figure 3
Figure 3:
Evolution of cycling peak power (CPP, ▪), time to reach CPP (tCPP, •), and optimal velocity (vopt, ▵) with age. ns, not statistically significant. ***P < 0.001

Differences between maximal power and PP at three selected FB.

For each variable studied, the relationship with age was calculated from the whole population. However, to simplify the presentation of the figures, only age groups with results obtained by contrast tests are represented. The difference between PP50 and CPP was constant with age (5–6%, Fig. 4b). For DPP25 and DPP75, it was verified that a third-order function provided slightly higher correlation than other functions. DPP25 remained steady until 15 yr and decreased after this age (for 19 yr, PP25 underestimated CPP by about 10%;Fig. 4a). Conversely, DPP75 increased with age and was close to zero after 16 yr (Fig. 4c).

Figure 4
Figure 4:
Differences in percent between cycling peak power (CPP) and peak power at a braking force of 0.245 N·kg\-1 body mass (PP25) (a), peak power at a braking force of 0.491 N·kg\-1 body mass (PP50) (b), and peak power at a braking force of 0.736 N·kg\-1 body mass (PP75) (c). ns, not statistically significant. ***P < 0.001.

Differences between vopt and three cycling velocities.

A significant relationship was found only between Dv75 and age (r = 0.47, P < 0.001). v75 underestimated vopt by more than 10% before 16 yr of age (Fig. 5c). Dv25 and Dv50 were close to zero in all age groups (Fig. 5a and 5b).

Figure 5
Figure 5:
Differences in percent between optimal velocity (vopt) and velocity at a braking force of 0.245 N·kg\-1 body mass (v25) (a), velocity at a braking force of 0.491 N·kg\-1 body mass (v50) (b), and velocity at a braking force of 0.736 N·kg\-1 body mass (v75) (c). ns, not statistically significant. ***P < 0.001.

Figure 6 shows that the time to attain PP at 0.736 N·kg−1BM (tPP75) was 6–8 s on average in children, while this time was only about 3 s in young adults.

Figure 6
Figure 6:
Evolution of time to reach peak power at braking force of 0.736 N·kg\-1 body mass (corresponding applied load of 75 g·kg\-1 body mass) (tPP75) vs age. ns, not statistically significant.


If “true” anaerobic power output has to be measured, the duration of the test must be as short as possible. Until now, the highest instantaneous maximal power reported was measured during a vertical jump test of 4 ms duration (7). The force-velocity test has the advantage of providing an accurate measure of power, force, and velocity components during short-term cycling (2). In adults, the method proposed by Lakomy (flywheel inertia included) enables the measurement of maximal mechanical power between 1 and 3 s of exercise (13,14,21). It is well documented that short-term cycling power increases with age (3,10,23). The young adults showed a CPP close to or slightly lower than previously reported data using similar methods on constant FB ergometers (1,13,14,21).

In this study, we found that PP was dependent on the FB. The difference between PP measured for a FB of 0.491 N·kg1 (corresponding braking load: 50 g·kg1) and the maximal value of CPP was constant with age. The cycling velocity, obtained in all age-groups, with a FB of 0.491 N·kg1 confirmed the latter data.

Young adults.

In young adults, PP for low FB underestimated CPP by more than 10%, whereas PP for high resistance was close to CPP. These results disagree with previous reported data (1,13,21) in which PP was found to be independent of FB. It is possible that the higher flywheel inertia in one study (1) (0.927 vs 0.546 kg·m2 in this study) masked the FB effects in the range of 25 and 55 N·kg1 BM. The total external resistive force (sum of frictional and inertial force) must be optimal, which allows the muscles to operate at their vopt. In this study, the adult population attained a cycling vopt of 130 rpm. The vopt values were close to those reported in the literature on a “constant” load cycle ergometer (125 rpm for Arsac et al. (1)) and more elevated than on a isokinetic cycle ergometer (20). For all FB used in this study, the cycling velocity at PP was close to the vopt. However, in adults, a FB of 0.245 N·kg1 BM was undoubtedly too low to obtain maximal power. In the method with flywheel inertia, it is not necessary to overcome resistances as high as in the “classical” method, i.e., higher than 100 N·kg1 BM in male adults (14,22). The present study showed that a FB lower than 0.736 N·kg1 BM was not high enough. Moreover, Linossier et al. (14), who studied the work performed to reach maximal velocity on the same type of cycle-ergometer (expressed in mean power output), determined an optimal FB of 0.844 N·kg1 BM (i.e., 86 g·kg1 BM).


In contrast to adults, a FB of 0.736 N·kg1 BM (corresponding applied load: 75 g·kg1 BM) was too high in children and PP75 underestimated CPP by about 15%. This observation was also confirmed by the values of cycling velocity at PP, which underestimated vopt by about 20% between 8 and 13 yr of age when this high resistance was used. The values of vopt (between 110 and 115 rpm for a FB of 0.245 N·kg1 or 0.491 N·kg1 BM) were close to those reported in the literature (112 rpm by Sargeant et al. (19) and 120 rpm by Sargeant and Dolan (20) on an isokinetic cycle ergometer or 114 rpm on a “constant” load cycle ergometer by Van Praagh et al. (25)). The present study showed that a FB higher than 0.491 N·kg1 BM in children does not elicit optimal cycling velocity and consequently does not attain the highest possible value for maximal power (see Fig. 1). This value is lower than the optimal FB recommended in previous studies that did not include flywheel inertia (3,6,24,25).

The optimal FB applied, even when normalized for BM, is therefore lower in children than in adults. It was hypothesized that the flywheel inertia was unadjusted for the prepubescent child (4). Developmental changes may also explain this difference. Malina and Bouchard (15) showed an increase of muscle mass, corrected for total BM, from about 42% to 54% in males from 5 to 17.5 yr of age. In the present study, we calculated the ratio between LLV and BM (LLV/BM): it was significantly higher in adults than in children (0.095 vs 0.075, respectively;P < 0.0001). Therefore, the use of BM as a ratio standard for FB seems irrelevant. LLV or lean BM are more suitable parameters when comparing age-related differences. Moreover, qualitative changes during growth such as fiber characteristics, hormonal effects, and neuromotor maturation were assumed to exert an important influence on the force-generating capacity of muscles (23).

It was also suggested that the lower velocities in children during high resistance, and consequently the lower CPP, was because of the onset of muscle fatigue (4). Our pediatric population needed more than 6 s to attain PP for a FB of 0.736 N·kg1 BM. In adults, the depletion of creatine phosphate stores after 5 to 7 s of maximal exercise leads to a reduction in the energy production rate and therefore a decrease in power output (9). Mercier et al. (17) demonstrated that blood lactate concentration increased after a single maximal sprint of 6 s in adults. This observation suggests that anaerobic glycolysis provides a part of the energy demand during a 6 s sprint. To our knowledge, no equivalent studies are available for children using a similar exercise duration. Using 31P nuclear magnetic resonance spectroscopy, Zanconato et al. (26) and Kuno et al. (12) showed that children (aged 7–10 yr) and adolescents (aged 12–15 yr) had a lower anaerobic metabolism during exercise compared with adults. Therefore, the depletion of high-energy phosphate stores and the subsequent reduction in energy production rate could be responsible for a reduction in maximal muscle work after 4 or 5 s of exercise in our pediatric population. Consequently, this may partly explain why they could not attain their optimal cycling velocity and produce the highest possible mechanical power during high FB.

The advantage of the method used in this study (flywheel inertia included) is to obtain values of CPP, vopt, and optimal force from a single sprint. In children, mass-related resistance is lower than in adults because of their inferior leg muscle force capabilities. For measurement of “true” CPP, the testing device must be adjusted with respect to the changes in body composition during growth. The present method is less affected by fatigue and motivational factors and is also less time consuming than current testing procedures. Therefore, using the actual cycle ergometer, one sprint with a single FB of 0.496 N·kg1 BM [corresponding applied load: 50 g·kg1 BM] is a relevant and feasible method to test heterogeneous male populations.

Nanci de França was supported by the CNPQ (Brazil).


1. Arsac, M. A., A. Belli, and J.-R. Lacour. Muscle function during brief maximal exercise: accurate measurements on a friction-loaded cycle ergometer. Eur. J. Appl. Physiol. 74:100–106, 1996.
2. Bar-Or, O. Anaerobic performance. In: Measurement in Pediatric Exercise Science. D. Docherty (Ed.). Champaign, IL: Human Kinetics, 1996, pp. 161–182.
3. Carlson, J. S. and G. A. Naughton. Performance characteristics of children using various braking resistances on the Wingate anaerobic test. J. Sports Med. Phys. Fitness 34:362–369, 1994.
4. Delgado, A., G. Pérès, A. Allemandou, and H. Monod. Influence of cycle ergometer characteristics on the adolescents anaerobic abilities testing. Arch. Int. Physiol. Bioch. Biophys. 101:145–148, 1993.
5. Doré, E., N. M. França, M. Bedu, and E. Van Praagh. The effect of flywheel inertia on short-term cycling power output in children. Med. Sci. Sports Exerc. 29:S170, 1997.
6. Dotan, R. and O. Bar-Or. Load optimization for the Wingate anaerobic test. Eur. J. Appl. Physiol. 51:409–417, 1983.
7. Ferretti, G., M. V. Narici, T. Binzoni, L. Gariod, J. F. Le Bas, H. Reutenauer, and P. Cerretelli. Determinants of peak muscle power: effects of age and physical conditioning. Eur. J. Appl. Physiol. 68:111–115, 1994.
8. Fouquet, R., A. Belli, J. Jay, et al. Measurement and exploitation of the power done on a cycle ergometer. Innov. Tech. Biol. Med. 14:709–717, 1993.
9. Hirvonen, J., S. Rhunen, H. Rusko, and M. Märkönen. Breakdown of high-energy phosphate compounds and lactate accumulation during short supramaximal exercise. Eur. J. Appl. Physiol. 56:253–259, 1987.
10. Inbar, O. and O. Bar-Or. Anaerobic characteristics in male children and adolescents. Med. Sci. Sports Exerc. 18:264–269, 1986.
11. Jones, P. R. and J. Pearson. Anthropometric determination of leg fat and muscle plus bone volumes in young male and females adults. J. Physiol. 204:63–66, 1969.
12. Kuno, S., H. Takahashi, K. Fujimoto, H. Akima, M. Miyamaru, I. Nemoto, Y. Itai, and S. Katsuta. Muscle metabolism during exercise using phosphorus-31 nuclear magnetic resonance spectroscopy in adolescents. Eur. J. Appl. Physiol. 70:301–304, 1995.
13. Lakomy, H. K. Measurement of work and power output using friction-loaded cycle ergometers. Ergonomics 29:509–517, 1986.
14. Linossier, M. T., D. Dormois, R. Fouquet, A. Geyssant, and C. Denis. Use of the force-velocity test to determine the optimal braking force for a sprint exercise on a friction-loaded cycle ergometer. Eur. J. Appl. Physiol. 74:420–427, 1996.
15. Malina, R. M. and C. Bouchard. Growth, maturation and physical activity. Champaign, IL: Human Kinetics, 1991, pp. 115–132.
16. Martin, J. C., B. M. Wagner, and E. F. Coyle. Inertial-load method determined maximal cycling power in a single exercise bout. Med. Sci. Sports Exerc. 29:1505–1512, 1997.
17. Mercier, B., P. Granier, J. Mercier, L. Foucat, G. Bielicki, J. Pradere, J. P. Renou, and C. Prefaut. Non-invasive skeletal muscle lactate detection between periods of intense exercise in humans. Eur. J. Appl. Physiol. 78:20–27, 1998.
18. Sargeant, A. J. The determinants of anaerobic function during growth. In: Pediatric Anaerobic Performance. E. Van Praagh (Ed.). Champaign, IL: Human Kinetics, 1998, pp. 97–117.
19. Sargeant, A. J., P. Dolan, and A. Young. Optimal velocity for maximal short-term (anaerobic) power output in cycling. Int. J. Sports Med. 5:(Suppl.)124–125, 1984.
20. Sargeant, A. and P. Dolan. Optimal velocity of muscle contraction for short-term (anaerobic) power output in children and adults. In:Children and Exercise, Vol. XII, J. Rutenfranz, R. Mocellin, and F. Klimt (Eds.). Champaign, IL: Human Kinetics, 1986, pp. 39–42.
21. Seck, D., H. Vandewalle, N. Decrops, and H. Monod. Maximal power and torque-velocity relationship on a single ergometer during the acceleration phase of a single all-out exercise. Eur. J. Appl. Physiol. 70:161–168, 1995.
22. Vandewalle, H., G. Peres, J. Heller, and H. Monod. All-out anaerobic capacity tests on cycle ergometers. Eur. J. Appl. Physiol. 54:222–229, 1985.
23. Van Praagh, E. Developmental aspects of anaerobic function. In:Children and Exercise, Vol. XIX, N. Armstrong, B. Kirby, and J. Welsman (Eds.). London: E & FN Spon, 1997, pp. 269–290.
24. Van Praagh, E., G. Falgairette, M. Bedu, N. Fellmann, and J. Coudert. Laboratory and field tests in 7-year-old boys. In:Children and Exercise, Vol. XIII, S. Oseid and K. H. Carlsen (Eds.). Champaign, IL: Human Kinetics, 1989, pp. 11–17.
25. Van Praagh, E., N. Fellmann, M. Bedu, G. Falgairette, and J. Coudert. Gender difference in the relationship of anaerobic power output to body composition in children. Pediatr. Exerc. Sci. 2:336–348, 1990.
26. Zanconato, S., S. Buchthal, T. J. Barstow, and D. M. Cooper.31P-magnetic resonance spectroscopy of leg muscle metabolism during exercise in children and adults. J. Appl. Physiol. 74:2214–2218, 1993.


©2000The American College of Sports Medicine