Energy expenditure
during bicycling
STEVE D. McCOLE, KEVIN CLANEY, JEAN-CLAUDE CONTE, RICHARD ANDERSON, AND JAMES M. HAGBERG Center for Exercise Science, Department of Exercise and Sport Science, College of Health and Human Performance, University of Florida, Gainesville, Florida 32611
MCCOLE, STEVE RICHARD ANDERSON,
D., KEVIN
CLANEY, JEAN-CLAUDE JAMES M. HAGBERG.
CONTE,
Energy expenditure during bicycling. J. Appl. Physiol. 68(2): 748-753, 1990.-This study was designed to measure the O2 uptake (vo2) of cyclists while they rode outdoors at speeds from 32 to 40 km/h. Regression analyses of data from 92 trials using the same wheels, tires, and tire pressure with the cyclists riding in their preferred gear and in an aerodynamic position indicated the best equation (r = 0.84) to estimate VO, in liters per minute vo* = -4.50
AND
+ 0.17 rider speed + 0.052 wind speed + 0.022 rider weight
where rider and wind speed are expressed in kilometers per hour and rider weight in kilograms. Following another rider closely, i.e., drafting, at 32 km/h reduced VO, by 18 t 11%; the benefit of drafting a single rider at 37 and 40 km/h was greater (27 t 8%) than that at 32 km/h. Drafting one, two, or four riders in a line at 40 km/h resulted in the same reduction in vo2 (27 k 7%). Riding at 40 km/h at the back of a group of eight riders reduced voz by significantly more (39 k 6%) than drafting one, two, or four riders in a line; drafting a vehicle at 40 km/h resulted in the greatest decrease in vo2 (62 * 6%). \jo2 was also 7 sf~4% lower when the cyclists were riding an aerodynamic bicycle. An aerodynamic set of wheels with a reduced number of spokes and one set of disk wheels were the only wheels to reduce %7oq significantly while the cyclists were riding a conventional racing bicycle at 40 km/h. Thus drafting and using aerodynamically designed equipment can alter the energy expenditure of cyclists at speeds similar to those encountered in competitive events (32-40 km/h).
that a cyclist encounters. Such alterations in air resistance can be achieved in competitive situations by following closely behind other riders, i.e., drafting, and by the use of technologically advanced equipment, especially wheels and bicycles, now available to cyclists. Using a system that we developed to measure cyclists’ Vo2 while they were riding outdoors, we sought to determine the energy cost of bicycling at speeds of 32-40 km/ h and to determine the effects of various techniques and equipment that alter the cyclists’ air resistance on their . vo 2* MATERIALS
AND
METHODS
Twenty-eight male competitive cyclists, all of whom were either registered US Cycling Federation riders or triathletes, volunteered for these experiments. The height and weight of the subjects averaged 181.3 t 7.1 cm (range 165.1-200.7 cm) and 73.5 t 7.7 kg (range 56.687.7 kg), respectively. They were riding 330 t 100 km/ wk in training at the time of the study. Their maximal O2 uptake, measured during a progressive test on their racing bicycles on a Schwinn Velodyne bicycling simulator, was 67 t 7 ml kg-’ min. The study had previously been approved by the University of Florida Institutional Review Board. Each subject performed all trials in the outdoor experiments while riding in a racing position on his own bicycle on a flat, straight road. A survey of the road indicated oxygen uptake; aerodynamics; air resistance that the total elevation change was ~2 m over the 2,000m section where expired gas collections occurred. Expired gases were collected in neoprene meteorological balloons THE ENERGY COST of running has been the subject of connected via a three-way valve and approximately 7m numerous investigations in the past 20 years (cf. Refs. of flexible tubing to a Daniels’ low-resistance breathing IO-12), and the results of these studies indicate that valve and mouthpiece. The tubing and breathing valve running speed and the athlete’s economy are the primary were supported by a 5-m fiberglass pole attached to a determinants of the energy expenditure. However, very rack on the roof of a following vehicle. The pole was little information exists on the energy cost of bicycling, positioned so that the mouthpiece, and hence the rider, perhaps because of the difficulty in measuring oxygen was approximately 0.3-1.0 m in front and 1 m to the consumption (Vo2) of cyclists riding outdoors. Most en- right of the following vehicle. The rider had a noseclip ergy expenditure measurements during bicycling have in place throughout the entire trial. This system is simbeen made at speeds ranging from 8 to 25 km/h (1, 4-6, ilar to that of Swain and co-workers (15) and others (13). 13,15,16), speeds at which rolling and mechanical forces All trials started with the cyclist riding for 4 min at provide most of the resistance to movement (1, 9, 17, the desired speed and condition without being connected 18). At bicycling speeds >25 km/h, the power required to the gas collection system. The cyclist then turned and to overcome air resistance exceeds that required to over- rode for 1.5 min before being connected to the gas colleccome both rolling and mechanical resistances. Thus, at tion system. Because the dead space of the system was these faster speeds, it is possible that energy expenditure roughly 6 liters, the expired gas collections were begun at least 45-60 s after the rider was connected to the could be markedly altered by changing the air resistance 748
l
0161-7567/90 $1.50 Copyright 0 1990 the American Physiological Society
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (165.134.107.087) on August 23, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
ENERGY
EXPENDITURE
system. Three 45 to 75-s collections of expired gases were made during the last 3 min of each trial. Thus gas collections were made when the rider was in minutes 911 of the steady-state trial; these collections were also always made over the same section of the course with the rider always riding in the same direction. The expired gas collections were analyzed in the laboratory l-3 h after they were collected. The O2 and CO2 contents of the collected gases were analyzed using a Perkin-Elmer MGA 1100 respiratory mass spectrometer. Analyses indicated that O2 fractions did not change over the l- to 3h time period between collection and analysis, whereas CO2 fractions changed < 1%. Volumes were measured on a Tissot chain-compensated spirometer. It was assumed, because the trials were 9-10 min long with the riders in a relatively steady-state condition, that anaerobic sources contributed minimally to total energy expenditure and that minimal variations in substrate utilization occurred in these trials. Throughout all trials, the cyclists rode with their hands on the dropped portion of the handlebars. During the expired gas collection period of each trial, wind plus rider speed was measured lo-15 times with a Dwyer Mark IV anemometer mounted on the roof of the following vehicle. The average speed of the rider during the collection period of each trial was measured with a computerized cyclometer; the cyclometer also provided the rider with continuous feedback of speed during the trial. Wind speed was calculated as the difference between the speed measured on the anemometer and the rider speed on the cyclometer. Trials were conducted in the early morning hours to minimize winds. The first outdoor experiment involved drafting one other rider at 32, 37, and 40 km/h; in addition, cyclists were studied while drafting two and four riders in a line, drafting at the back of a group of eight riders, and drafting a vehicle at 40 km/h. The eight-rider group had two riders in the front row, three in the second row, and the subject riding between two other riders in the third row, i.e., pack formation. In these trials, the cyclist being studied rode 0.2-0.5 m from the vehicle or the rear wheel of the rider in front of him. These conditions were maintained throughout the entire 9- to ll-min trial. The second outdoor experiment consisted of cyclists riding alone at 40 km/h using different sets of racing wheels (Table 1) or an aerodynamic bicycle. The aerodynamic bicycle had cowhorn handlebars and a sloping top tube and used a 24in. front wheel and the rear wheel of the second set of disk wheels (Table 1); without wheels, it weighed somewhat less (6.4 vs. 6.6-7.0 kg) than the conventional bicycles used in these studies. The different wheel sets were also studied in the laboratory while the cyclists were riding at 40 km/h on self-driven rollers. Control trials for all outdoor experiments were conducted with the cyclist riding alone at the desired speed using a standard set of wheels (Table 1). The data from these trials were used in subsequent regression analyses to determine the effects of rider speed, wind speed, and rider weight on \jo2 during bicycling. Tire pressure was 100 psi for all trials. Tires on all wheels were 280-g slick sew-up racing tires. All wheels were fit with the same
DURING
BICYCLING
749
gears. Each rider selected his preferred gear ratio, and the gear remained the same for all trials at that particular speed (7). Each cyclist also wore the same helmet and clothing for all trials within a study. Trials within a study were completed in a random order, and lo-20 min of recovery were allowed between trials depending on the relative intensity for each cyclist. The data are presented as means t SD. The effects of rider speed, wind speed, and rider weight on energy expenditure were determined using uni- and multivariate regression analysis techniques within the Statistical Analysis System (2). The statistical significance of differences was assessed by testing appropriate subhypotheses within an analysis of covariance framework with rider speed and wind speed as covariates (2). RESULTS %Jo, averaged 2.59 t 0.44 (n = ll), 3.59 t 0.38 (n = l2), and 4.03 t 0.41 (n = 69) l/min while the cyclists were riding at 32, 37, and 40 km/h, respectively (Fig. 1); when expressed relative to body weight Vo2 averaged 37.4 t 5.9, 51.2 t 5.8, and 55.9 t 6.6 mlokg-lomin-l at these three speeds. Ninety-two trials were completed using the same wheels, tires, and tire pressure with cyclists in their preferred gear and riding in an aerodynamic position. When these data were subjected to regression analyses, in general the independent variables best predicted the Vo2 value expressed in liters per minute (Table 2), and the single factor having the greatest predictive capacity for Vo2 in liters per minute was rider speed (r = 0.73). Although wind speed and rider weight generally correlated independently with Tjo2 during these trials, they accounted for only 5-7% of the variance in TO, (Table 2). The best equation to estimate vo2 included wind speed and rider weight along with rider speed (r = 0.84) . vo 2= -4.50 + 0.17 VR + 0.052 VW + 0.022 WR
where VOW is expressed in liters per minute, VR is rider speed in kilometers per hour, VW is wind speed in kilometers per hour, and WR is rider weight in kilograms. At 32 km/h, $702 was reduced by 18 t 11% when the cyclists were drafting a single rider compared with when riding alone (Fig. 2). The reduction in VOW while the cyclists were drafting a single rider at 37 and 40 km/h was 28 t 10 and 26 t 8%, respectively, which was both significantly larger than the effect of drafting a single rider at 32 km/h but not different from each other. Riding at 40 km/h, drafting one rider, or a line of two or four riders all resulted in the same reduction in vo2 (27 -+ 7%, Fig. 3). However, riding at the back of the eightrider pack formation at 40 km/h reduced VOW by significantly more (39 t 6%) than riding behind one, two, or four riders in a line. The greatest decrease in Vo2 during these studies (62 t 6%) occurred when drafting the vehicle at 40 km/h (Fig. 3). Ventilation (VE) and respiratory exchange ratio (RER) decreased as a function of the reduction in VO,; when VOWwas lower, in most cases both VE and RER were also reduced significantly. Two sets of wheels resulted in significant reductions
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (165.134.107.087) on August 23, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
750
ENERGY
EXPENDITURE
DURING
BICYCLING
1. Descriptive information on wheel combinations used in outdoor and laboratory studies
TABLE
Wheel Wheels
Front
Control
Weight,
Description
wheels
Campagnolo hub with 32 double-butted Mavic GEL 280 rims
spokes,
kg Rear
wheel
wheel
1.025
1.550
Aerodynamic
16-18
Roval hubs, bladed rear, aerodynamic
spokes, 16-spoke front, l&spoke rims, recessed spoke nipples
1.075
1.525
Aerodynamic
24-32
Roval hubs, bladed rear, aerodynamic
spokes, 24-spoke front, 3%spoke rims, recessed spoke nipples
1.110
1.525
*
3.250
*
1.675
1.350
1.740
1.625
1.900
0.850
t
Rear disk 1
Ambrosio
Rear disk 2
Same as control rear wheel except spokes by urethane-coated nylon (Unidisk)
Double
disk 1
Epoxy composite disk bonded with sealed hubs (HED)
Double
disk 2
Flat composite honeycomb fiber sheets, spun carbon hubs, 26-in. front wheel, (Wolber Record Discjets)
24-in.
front
Radially
smooth
spoked
enclosed
front
wheel
rear disk wheel covered
to aluminum
rims
core between carbon fiber rims, and DuraAce 2%in. rear wheel
for aerodynamic
bike
All weights were determined with tires on wheels and the same freewheel on all rear wheels. * Control front wheel was used in combination with rear disk wheels. t Rear wheel of double disk 2 set of wheels described above was used in studies on aerodynamic bike.
6
TABLE 2. Univariate and multivariate regression equations derived to predict i'oz during bicycling
F
Equation
Swain Pugh McCole
10
20
30
40
50
Speed (km/hr) FIG. 1. Relationship between voq and riding speed from data in the present and previous studies. Pugh (13) measured \i02 of 6 cyclists at speeds of 32 and 37 km/h and of 3 cyclists at 40 km/h. Data from Swain et al. (15) are presented for speeds of 16-32 km/h. Body weights of the cyclists in the studies of Pugh and Swain et al. were similar to those in the present study. Solid lines represent values estimated by Whitt (17) from theoretical calculations corrected, using weight correction for voz derived in the present study, to body weights of riders in the present study and those of Pugh (13) and Swain et al. (15).
in voz when the cyclists rode a conventional racing bicycle alone at 40 km/h (Fig. 4); the 16- to M-spoke aerodynamic wheels reduced VOW by 7 t 5%, and the first double disk set of wheels reduced 90~ by 3 t 4%. The reduction in Vo2 resulting from the second rear disk wheel set approached significance (4 t 5%, P = 0.06). Vo2 was also significantly lower, by 7 t 4%, when the cyclists were riding the aerodynamic bicycle at 40 km/h. Again VE and RER differences were generally the same
Equations vo2 = VOf = vo, = VO2 = VO2 = VO, = iroz =
to predict vo2 in ml. kg-’ min-’ -26.62 + 2.04 VR 52.60 + 0.28 VW 72.95 - 0.28 WR -34.84 + 2.24 VR + 0.75 VW -3.24 + 2.22 VR - 0.42 WR 67.30 + 0.50 VW - 0.20 WR -16.30 + 2.37 VR + 0.64 VW - 0.33
Equations vo, = 90, = VO, = 60, =
predict VO, in 1. min-’ + 0.17 VR + 0.030 VW + 0.024 WR + 0.18 VR + 0.044 VW + 0.16 VR + 0.013 WR + 0.042 VW + 0.030 WR + 0.17 VR + 0.052 VW +
to -2.65 3.77 2.03 -3.15 -3.44 vo, = 1.56 VO, = -4.50
voz=
VR, rider weight
l
0.53” 0.26.f 0.23.t 0.75’ 0.72* 0.32t 0.80*
WR
0.73* 0.20$ 0.28t 0.81* 0.74’ 0.39* 0.84*
0.022 WR
velocity in km/h; VW, wind velocity in kg. * P < 0.01; t P < 0.05; $ P = 0.07.
in km/h;
WR, rider
as those for Vo2. The only reduction in VOW elicited by the different wheel combinations in the laboratory on self-driven rollers was a 4 t 4% reduction in VO, elicited with the first double disk set of wheels (data not shown). DISCUSSION
Numerous factors influence the forces that oppose movement in bicycling. At speeds ~20-25 km/h, rolling and mechanical resistances provide the majority of the opposition to movement (1,9,17,18). However, at speeds >25 km/h, air resistance provides the major obstacle to movement. The effect of these resistances can be demonstrated by comparing how they change across a range
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (165.134.107.087) on August 23, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
ENERGY
EXPENDITURE
DURING
N14 0 I>
32
Speed (km/hr) FIG.
different 0.05).
2. Reduction in %‘OZ resulting from drafting speeds. *Significantly different from that
a single rider at 3 at 32 km/h (P
during bicycling
STEVE D. McCOLE, KEVIN CLANEY, JEAN-CLAUDE CONTE, RICHARD ANDERSON, AND JAMES M. HAGBERG Center for Exercise Science, Department of Exercise and Sport Science, College of Health and Human Performance, University of Florida, Gainesville, Florida 32611
MCCOLE, STEVE RICHARD ANDERSON,
D., KEVIN
CLANEY, JEAN-CLAUDE JAMES M. HAGBERG.
CONTE,
Energy expenditure during bicycling. J. Appl. Physiol. 68(2): 748-753, 1990.-This study was designed to measure the O2 uptake (vo2) of cyclists while they rode outdoors at speeds from 32 to 40 km/h. Regression analyses of data from 92 trials using the same wheels, tires, and tire pressure with the cyclists riding in their preferred gear and in an aerodynamic position indicated the best equation (r = 0.84) to estimate VO, in liters per minute vo* = -4.50
AND
+ 0.17 rider speed + 0.052 wind speed + 0.022 rider weight
where rider and wind speed are expressed in kilometers per hour and rider weight in kilograms. Following another rider closely, i.e., drafting, at 32 km/h reduced VO, by 18 t 11%; the benefit of drafting a single rider at 37 and 40 km/h was greater (27 t 8%) than that at 32 km/h. Drafting one, two, or four riders in a line at 40 km/h resulted in the same reduction in vo2 (27 k 7%). Riding at 40 km/h at the back of a group of eight riders reduced voz by significantly more (39 k 6%) than drafting one, two, or four riders in a line; drafting a vehicle at 40 km/h resulted in the greatest decrease in vo2 (62 * 6%). \jo2 was also 7 sf~4% lower when the cyclists were riding an aerodynamic bicycle. An aerodynamic set of wheels with a reduced number of spokes and one set of disk wheels were the only wheels to reduce %7oq significantly while the cyclists were riding a conventional racing bicycle at 40 km/h. Thus drafting and using aerodynamically designed equipment can alter the energy expenditure of cyclists at speeds similar to those encountered in competitive events (32-40 km/h).
that a cyclist encounters. Such alterations in air resistance can be achieved in competitive situations by following closely behind other riders, i.e., drafting, and by the use of technologically advanced equipment, especially wheels and bicycles, now available to cyclists. Using a system that we developed to measure cyclists’ Vo2 while they were riding outdoors, we sought to determine the energy cost of bicycling at speeds of 32-40 km/ h and to determine the effects of various techniques and equipment that alter the cyclists’ air resistance on their . vo 2* MATERIALS
AND
METHODS
Twenty-eight male competitive cyclists, all of whom were either registered US Cycling Federation riders or triathletes, volunteered for these experiments. The height and weight of the subjects averaged 181.3 t 7.1 cm (range 165.1-200.7 cm) and 73.5 t 7.7 kg (range 56.687.7 kg), respectively. They were riding 330 t 100 km/ wk in training at the time of the study. Their maximal O2 uptake, measured during a progressive test on their racing bicycles on a Schwinn Velodyne bicycling simulator, was 67 t 7 ml kg-’ min. The study had previously been approved by the University of Florida Institutional Review Board. Each subject performed all trials in the outdoor experiments while riding in a racing position on his own bicycle on a flat, straight road. A survey of the road indicated oxygen uptake; aerodynamics; air resistance that the total elevation change was ~2 m over the 2,000m section where expired gas collections occurred. Expired gases were collected in neoprene meteorological balloons THE ENERGY COST of running has been the subject of connected via a three-way valve and approximately 7m numerous investigations in the past 20 years (cf. Refs. of flexible tubing to a Daniels’ low-resistance breathing IO-12), and the results of these studies indicate that valve and mouthpiece. The tubing and breathing valve running speed and the athlete’s economy are the primary were supported by a 5-m fiberglass pole attached to a determinants of the energy expenditure. However, very rack on the roof of a following vehicle. The pole was little information exists on the energy cost of bicycling, positioned so that the mouthpiece, and hence the rider, perhaps because of the difficulty in measuring oxygen was approximately 0.3-1.0 m in front and 1 m to the consumption (Vo2) of cyclists riding outdoors. Most en- right of the following vehicle. The rider had a noseclip ergy expenditure measurements during bicycling have in place throughout the entire trial. This system is simbeen made at speeds ranging from 8 to 25 km/h (1, 4-6, ilar to that of Swain and co-workers (15) and others (13). 13,15,16), speeds at which rolling and mechanical forces All trials started with the cyclist riding for 4 min at provide most of the resistance to movement (1, 9, 17, the desired speed and condition without being connected 18). At bicycling speeds >25 km/h, the power required to the gas collection system. The cyclist then turned and to overcome air resistance exceeds that required to over- rode for 1.5 min before being connected to the gas colleccome both rolling and mechanical resistances. Thus, at tion system. Because the dead space of the system was these faster speeds, it is possible that energy expenditure roughly 6 liters, the expired gas collections were begun at least 45-60 s after the rider was connected to the could be markedly altered by changing the air resistance 748
l
0161-7567/90 $1.50 Copyright 0 1990 the American Physiological Society
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (165.134.107.087) on August 23, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
ENERGY
EXPENDITURE
system. Three 45 to 75-s collections of expired gases were made during the last 3 min of each trial. Thus gas collections were made when the rider was in minutes 911 of the steady-state trial; these collections were also always made over the same section of the course with the rider always riding in the same direction. The expired gas collections were analyzed in the laboratory l-3 h after they were collected. The O2 and CO2 contents of the collected gases were analyzed using a Perkin-Elmer MGA 1100 respiratory mass spectrometer. Analyses indicated that O2 fractions did not change over the l- to 3h time period between collection and analysis, whereas CO2 fractions changed < 1%. Volumes were measured on a Tissot chain-compensated spirometer. It was assumed, because the trials were 9-10 min long with the riders in a relatively steady-state condition, that anaerobic sources contributed minimally to total energy expenditure and that minimal variations in substrate utilization occurred in these trials. Throughout all trials, the cyclists rode with their hands on the dropped portion of the handlebars. During the expired gas collection period of each trial, wind plus rider speed was measured lo-15 times with a Dwyer Mark IV anemometer mounted on the roof of the following vehicle. The average speed of the rider during the collection period of each trial was measured with a computerized cyclometer; the cyclometer also provided the rider with continuous feedback of speed during the trial. Wind speed was calculated as the difference between the speed measured on the anemometer and the rider speed on the cyclometer. Trials were conducted in the early morning hours to minimize winds. The first outdoor experiment involved drafting one other rider at 32, 37, and 40 km/h; in addition, cyclists were studied while drafting two and four riders in a line, drafting at the back of a group of eight riders, and drafting a vehicle at 40 km/h. The eight-rider group had two riders in the front row, three in the second row, and the subject riding between two other riders in the third row, i.e., pack formation. In these trials, the cyclist being studied rode 0.2-0.5 m from the vehicle or the rear wheel of the rider in front of him. These conditions were maintained throughout the entire 9- to ll-min trial. The second outdoor experiment consisted of cyclists riding alone at 40 km/h using different sets of racing wheels (Table 1) or an aerodynamic bicycle. The aerodynamic bicycle had cowhorn handlebars and a sloping top tube and used a 24in. front wheel and the rear wheel of the second set of disk wheels (Table 1); without wheels, it weighed somewhat less (6.4 vs. 6.6-7.0 kg) than the conventional bicycles used in these studies. The different wheel sets were also studied in the laboratory while the cyclists were riding at 40 km/h on self-driven rollers. Control trials for all outdoor experiments were conducted with the cyclist riding alone at the desired speed using a standard set of wheels (Table 1). The data from these trials were used in subsequent regression analyses to determine the effects of rider speed, wind speed, and rider weight on \jo2 during bicycling. Tire pressure was 100 psi for all trials. Tires on all wheels were 280-g slick sew-up racing tires. All wheels were fit with the same
DURING
BICYCLING
749
gears. Each rider selected his preferred gear ratio, and the gear remained the same for all trials at that particular speed (7). Each cyclist also wore the same helmet and clothing for all trials within a study. Trials within a study were completed in a random order, and lo-20 min of recovery were allowed between trials depending on the relative intensity for each cyclist. The data are presented as means t SD. The effects of rider speed, wind speed, and rider weight on energy expenditure were determined using uni- and multivariate regression analysis techniques within the Statistical Analysis System (2). The statistical significance of differences was assessed by testing appropriate subhypotheses within an analysis of covariance framework with rider speed and wind speed as covariates (2). RESULTS %Jo, averaged 2.59 t 0.44 (n = ll), 3.59 t 0.38 (n = l2), and 4.03 t 0.41 (n = 69) l/min while the cyclists were riding at 32, 37, and 40 km/h, respectively (Fig. 1); when expressed relative to body weight Vo2 averaged 37.4 t 5.9, 51.2 t 5.8, and 55.9 t 6.6 mlokg-lomin-l at these three speeds. Ninety-two trials were completed using the same wheels, tires, and tire pressure with cyclists in their preferred gear and riding in an aerodynamic position. When these data were subjected to regression analyses, in general the independent variables best predicted the Vo2 value expressed in liters per minute (Table 2), and the single factor having the greatest predictive capacity for Vo2 in liters per minute was rider speed (r = 0.73). Although wind speed and rider weight generally correlated independently with Tjo2 during these trials, they accounted for only 5-7% of the variance in TO, (Table 2). The best equation to estimate vo2 included wind speed and rider weight along with rider speed (r = 0.84) . vo 2= -4.50 + 0.17 VR + 0.052 VW + 0.022 WR
where VOW is expressed in liters per minute, VR is rider speed in kilometers per hour, VW is wind speed in kilometers per hour, and WR is rider weight in kilograms. At 32 km/h, $702 was reduced by 18 t 11% when the cyclists were drafting a single rider compared with when riding alone (Fig. 2). The reduction in VOW while the cyclists were drafting a single rider at 37 and 40 km/h was 28 t 10 and 26 t 8%, respectively, which was both significantly larger than the effect of drafting a single rider at 32 km/h but not different from each other. Riding at 40 km/h, drafting one rider, or a line of two or four riders all resulted in the same reduction in vo2 (27 -+ 7%, Fig. 3). However, riding at the back of the eightrider pack formation at 40 km/h reduced VOW by significantly more (39 t 6%) than riding behind one, two, or four riders in a line. The greatest decrease in Vo2 during these studies (62 t 6%) occurred when drafting the vehicle at 40 km/h (Fig. 3). Ventilation (VE) and respiratory exchange ratio (RER) decreased as a function of the reduction in VO,; when VOWwas lower, in most cases both VE and RER were also reduced significantly. Two sets of wheels resulted in significant reductions
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (165.134.107.087) on August 23, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
750
ENERGY
EXPENDITURE
DURING
BICYCLING
1. Descriptive information on wheel combinations used in outdoor and laboratory studies
TABLE
Wheel Wheels
Front
Control
Weight,
Description
wheels
Campagnolo hub with 32 double-butted Mavic GEL 280 rims
spokes,
kg Rear
wheel
wheel
1.025
1.550
Aerodynamic
16-18
Roval hubs, bladed rear, aerodynamic
spokes, 16-spoke front, l&spoke rims, recessed spoke nipples
1.075
1.525
Aerodynamic
24-32
Roval hubs, bladed rear, aerodynamic
spokes, 24-spoke front, 3%spoke rims, recessed spoke nipples
1.110
1.525
*
3.250
*
1.675
1.350
1.740
1.625
1.900
0.850
t
Rear disk 1
Ambrosio
Rear disk 2
Same as control rear wheel except spokes by urethane-coated nylon (Unidisk)
Double
disk 1
Epoxy composite disk bonded with sealed hubs (HED)
Double
disk 2
Flat composite honeycomb fiber sheets, spun carbon hubs, 26-in. front wheel, (Wolber Record Discjets)
24-in.
front
Radially
smooth
spoked
enclosed
front
wheel
rear disk wheel covered
to aluminum
rims
core between carbon fiber rims, and DuraAce 2%in. rear wheel
for aerodynamic
bike
All weights were determined with tires on wheels and the same freewheel on all rear wheels. * Control front wheel was used in combination with rear disk wheels. t Rear wheel of double disk 2 set of wheels described above was used in studies on aerodynamic bike.
6
TABLE 2. Univariate and multivariate regression equations derived to predict i'oz during bicycling
F
Equation
Swain Pugh McCole
10
20
30
40
50
Speed (km/hr) FIG. 1. Relationship between voq and riding speed from data in the present and previous studies. Pugh (13) measured \i02 of 6 cyclists at speeds of 32 and 37 km/h and of 3 cyclists at 40 km/h. Data from Swain et al. (15) are presented for speeds of 16-32 km/h. Body weights of the cyclists in the studies of Pugh and Swain et al. were similar to those in the present study. Solid lines represent values estimated by Whitt (17) from theoretical calculations corrected, using weight correction for voz derived in the present study, to body weights of riders in the present study and those of Pugh (13) and Swain et al. (15).
in voz when the cyclists rode a conventional racing bicycle alone at 40 km/h (Fig. 4); the 16- to M-spoke aerodynamic wheels reduced VOW by 7 t 5%, and the first double disk set of wheels reduced 90~ by 3 t 4%. The reduction in Vo2 resulting from the second rear disk wheel set approached significance (4 t 5%, P = 0.06). Vo2 was also significantly lower, by 7 t 4%, when the cyclists were riding the aerodynamic bicycle at 40 km/h. Again VE and RER differences were generally the same
Equations vo2 = VOf = vo, = VO2 = VO2 = VO, = iroz =
to predict vo2 in ml. kg-’ min-’ -26.62 + 2.04 VR 52.60 + 0.28 VW 72.95 - 0.28 WR -34.84 + 2.24 VR + 0.75 VW -3.24 + 2.22 VR - 0.42 WR 67.30 + 0.50 VW - 0.20 WR -16.30 + 2.37 VR + 0.64 VW - 0.33
Equations vo, = 90, = VO, = 60, =
predict VO, in 1. min-’ + 0.17 VR + 0.030 VW + 0.024 WR + 0.18 VR + 0.044 VW + 0.16 VR + 0.013 WR + 0.042 VW + 0.030 WR + 0.17 VR + 0.052 VW +
to -2.65 3.77 2.03 -3.15 -3.44 vo, = 1.56 VO, = -4.50
voz=
VR, rider weight
l
0.53” 0.26.f 0.23.t 0.75’ 0.72* 0.32t 0.80*
WR
0.73* 0.20$ 0.28t 0.81* 0.74’ 0.39* 0.84*
0.022 WR
velocity in km/h; VW, wind velocity in kg. * P < 0.01; t P < 0.05; $ P = 0.07.
in km/h;
WR, rider
as those for Vo2. The only reduction in VOW elicited by the different wheel combinations in the laboratory on self-driven rollers was a 4 t 4% reduction in VO, elicited with the first double disk set of wheels (data not shown). DISCUSSION
Numerous factors influence the forces that oppose movement in bicycling. At speeds ~20-25 km/h, rolling and mechanical resistances provide the majority of the opposition to movement (1,9,17,18). However, at speeds >25 km/h, air resistance provides the major obstacle to movement. The effect of these resistances can be demonstrated by comparing how they change across a range
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (165.134.107.087) on August 23, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
ENERGY
EXPENDITURE
DURING
N14 0 I>
32
Speed (km/hr) FIG.
different 0.05).
2. Reduction in %‘OZ resulting from drafting speeds. *Significantly different from that
a single rider at 3 at 32 km/h (P
