Bicycle Shock Absorption Systems and Energy Expended by the Cyclist

Abstract

Bicycle suspension systems have been designed to improve bicycle comfort and handling by dissipating terrain-induced energy. However, they may also dissipate the cyclist’s energy through small oscillatory movements, often termed ‘bobbing’, that are generated by the pedalling movements. This phenomenon is a major concern for competitive cyclists engaged in events where most of the time is spent climbing, e.g. off-road cross-country races. An acceptable method to assess the overall efficacy of suspension systems would be to evaluate energy consumed by cyclists using different types of suspension systems. It could be assumed that any system that reduces metabolic expenditure for the cyclist would automatically lead to performance improvement. Unfortunately, only a limited number of studies have been conducted on that subject. Moreover, the conclusions that can be drawn from most of them are limited due to unsatisfactory statistical power, experimental protocols, measuring techniques and equipment.
This review presents and discusses the most relevant results of studies that focused on mechanical simulations as well as on energy expenditure in relation to off-road bicycle suspension systems.
Evidence in the literature suggests that cyclist-generated power that is dissipated by suspensions is minimal and probably negligible on most terrains. However, the scarce studies on the topic as well as the limitations in the conclusions that can be drawn from most of them indicate that we should remain cautious before supporting the use of dual suspension bicycles on all course types and for all cyclists. For example, it should be kept in mind that most cross-country racers still use front suspension bicycles. This might be explained by excessive cyclist-generated power dissipation at the high mechanical powers developed by elite cross-country cyclists that have not been studied in the literature.
Finally, suspended bicycles are more comfortable. Moreover, the fact that suspension systems may significantly reduce physical stress should not be overlooked, especially in very long events and for recreational cyclists.

Full text

Sports Med 2004; 34 (2): 71-80
L
EADING
A
RTICLE
0112-1642/04/0002-0071/$31.00/0
 2004 Adis Data Information BV. All rights reserved
Bicycle Shock Absorption Systems
and Energy Expended by the Cyclist
Henri Nielens and Thierry Lejeune
Saint-Luc University Hospital, Catholic University of Louvain, Brussels, Belgium
Bicycle suspension systems have been designed to improve bicycle comfort
Abstract
and handling by dissipating terrain-induced energy. However, they may also
dissipate the cyclist’s energy through small oscillatory movements, often termed
‘bobbing’, that are generated by the pedalling movements. This phenomenon is a
major concern for competitive cyclists engaged in events where most of the time
is spent climbing, e.g. off-road cross-country races. An acceptable method to
assess the overall efficacy of suspension systems would be to evaluate energy
consumed by cyclists using different types of suspension systems. It could be
assumed that any system that reduces metabolic expenditure for the cyclist would
automatically lead to performance improvement. Unfortunately, only a limited
number of studies have been conducted on that subject. Moreover, the conclusions
that can be drawn from most of them are limited due to unsatisfactory statistical
power, experimental protocols, measuring techniques and equipment.
This review presents and discusses the most relevant results of studies that
focused on mechanical simulations as well as on energy expenditure in relation to
off-road bicycle suspension systems.
Evidence in the literature suggests that cyclist-generated power that is dissipat-
ed by suspensions is minimal and probably negligible on most terrains. However,
the scarce studies on the topic as well as the limitations in the conclusions that can
be drawn from most of them indicate that we should remain cautious before
supporting the use of dual suspension bicycles on all course types and for all
cyclists. For example, it should be kept in mind that most cross-country racers still
use front suspension bicycles. This might be explained by excessive cyclist-
generated power dissipation at the high mechanical powers developed by elite
cross-country cyclists that have not been studied in the literature.
Finally, suspended bicycles are more comfortable. Moreover, the fact that
suspension systems may significantly reduce physical stress should not be over-
looked, especially in very long events and for recreational cyclists.
Since their introduction in the US in the early cently, shock absorption systems also known as
1980s, mountain bikes have become increasingly ‘suspension systems’ or more simply ‘suspensions’
popular. In the beginning, off-road riding capabili- have been developed to improve comfort and per-
ties and gear shifting levers located on the handlebar formance. Nowadays, mountain bikes are the num-
essentially contributed to their popularity. More re- ber-one sold bicycle in the US and in Europe. Most

72 Nielens & Lejeune
models are equipped with a suspended front wheel over, Wang and Hull
[8,9]
emphasised that suspen-
sions may dissipate the power generated by the
often called ‘front suspension’ (FS). Some come
cyclist, which may become unacceptable for cyclists
with suspension systems on both wheels generally
participating in events organised on hilly terrain.
known as dual suspension (DS) systems.
This issue will be discussed more extensively in
When it seems obvious that suspensions improve
section 1. Power dissipation may occur in the sus-
comfort, only a limited number of studies have
pensions themselves or in the articulated bicycle
investigated the effects of such systems on the
frame that becomes more flexible,
[10]
especially in
mechanics and the energetics of cycling. After a
bicycles equipped with DS.
brief preliminary discussion of some relevant issues
concerning bicycle suspensions, this paper will re-
1. Bicycle Suspension Systems
view and discuss available data on the effects of
In general, most front and rear shock absorbing
bicycle shock absorption systems on energy expen-
devices are composed of an elastic and a viscous
diture.
element mounted in parallel (figure 1). Mechanical
As discussed by De Lorenzo et al.,
[1]
suspensions
properties of both elements are generally separately
isolate the cyclist from vibrations
[1]
and terrain-
adjustable on most bicycles. The elastic element is
induced shocks
[2,3]
by allowing the wheels to move
made of a steel spring that can be pre-constrained at
independently versus the rest of the bicycle. The
different levels or an air chamber that can be pre-
cyclist and the bicycle equipped with suspensions
inflated at varied pressures according to the nature
are therefore able to travel a smoother path as only
of the terrain and the cyclist’s preference. The vis-
the wheels follow the contours of the terrain, im-
cous element is generally made of a piston and
proving comfort and bicycle handling.
[2-7]
Suspen-
cylinder chamber filled with oil. The oil travels
sions may also improve cornering,
[2,3]
braking capa-
through orifices made in the piston. The total size of
city,
[2,5]
and more generally, bicycle control, han-
the orifices may be adjusted to modify the damper
dling
[4,7]
and traction
[2,5]
since they allow better
viscosity. Some simpler and cheaper systems in-
contact between the tyres and the ground. These
clude an elastomer part that has both viscous and
numerous advantages explain why all downhill
mountain bike racers use front and rear large travel
suspension systems that allow much higher speeds
in downhill events.
Alternatively, bicycle suspensions also have
drawbacks. For example, DS bicycles are much
more expensive and average between 1–2kg and
even 3kg heavier than equivalent rigid models. Such
a 10–15% increase in bicycle weight may be a
concern during uphill racing and accelerations as it
will demand an excess of energy expenditure for the
competitor. Wang and Hull
[8]
calculated that a 1.8kg
increase in bicycle weight would result in an extra
4.7 seconds for a 60kg cyclist to climb a 6% grade
hill for 1000m at 6.5 m/sec. On the basis of such
data, Wang and Hull calculated that such an appar-
ently small weight increase would lead to a total
time increase of 46 seconds on a championship race
similar to the Women’s World Mountain Bike
Championships held in Germany in 1995. More-
To wheel, ground
To rider, bike frame
Viscous
element
(damper)
Elastic
element
(spring)
Fig. 1. Components of a shock-absorbing device. Most shock-ab-
sorbing devices that equip modern suspended bicycles are com-
posed of an elastic (spring) and a viscous (damper) element mount-
ed in parallel between the wheel and the frame of the bicycle.
Mechanical properties of each element can generally be tuned
separately to adjust for terrain characteristics and cyclist prefer-
ences.
 2004 Adis Data Information BV. All rights reserved Sports Med 2004; 34 (2)

Bicycle Suspensions and Energy Expenditure 73
b
c
d
e
f
a
Suspension element
Rear suspension pivot
Fig. 2. Most common bicycle suspension designs: (a) front telescopic visco-elastic suspension system integrated in each arm of the fork; (b)
traditional rear swing arm with one pivot point; (c) multi-bar linkage rear suspension systems with multiple pivot points; (d) unified rear
triangle system that eliminates interactions between front chain-ring and rear suspension; (e) the Allsop’s Softride system; and (f)
suspension seat posts.
elastic properties. On some bicycles, the suspension bar linkage rear suspension systems with multiple
system may be turned off through command switch-
pivot points (figure 2c); and a unified rear triangle
es located on the handlebar allowing riding in rigid
system that eliminates interactions between the front
mode.
chain-ring and the rear suspension (figure 2d). Much
simpler systems have also been developed that allow
While most FS systems are generally made from
isolating the cyclist from vibrations and/or shocks
telescopic forks with visco-elastic elements in each
generated by terrain irregularities. However, in such
arm of the fork, rear suspension systems are numer-
systems, almost the entire bicycle mass is unsprung.
ous. Figure 2 shows several systems that are com-
In the Allsop’s Softride system, the saddle is mount-
monly available on the market. Suspension systems
ed on a flexible composite beam (figure 2e). Suspen-
are varied, from the simplest to the most sophisticat-
ed. Many mountain bikes are currently equipped sion seat posts are relatively cheap and simple sys-
only with a front telescopic visco-elastic suspension tems that can replace traditional rigid seat posts on
system integrated in each arm of the fork (figure 2a). most bicycles (figure 2f). The location of the centre
More sophisticated DS systems are: traditional rear of rotation of the rear suspension swing-arm (i.e. the
swing arm with one pivot point (figure 2b); multi- pivot point) is an important technical characteristic
 2004 Adis Data Information BV. All rights reserved Sports Med 2004; 34 (2)

74 Nielens & Lejeune
of rear suspensions as it may influence the magni- while noting that both the tyres
[2,8,10,15]
and the body
tude of cyclist-generated power that is dissipated in parts of the cyclist
[2,3]
also dissipate energy generat-
the system.
[8,11-14]
ed by terrain irregularities and thus act as dampers.
Downhill racers are confronted with important ter-
Several authors have studied the optimal pivot
rain irregularities on very steep courses travelled at
point location of the rear suspension through
high speeds. Hence, suspensions used in such events
mechanical modelling
[8,9,13]
and experimentation.
[3]
must be designed to absorb a large amount of ter-
However, only very few scientific publications have
rain-induced energy. As a result, modern downhill
addressed the issue of the optimal design for suspen-
bicycles are usually equipped with large travel sus-
sion systems. Accordingly, besides more precise
pensions with large energy absorption capacity.
knowledge about rear suspension pivot point loca-
tion that will be presented and discussed more ex-
Unfortunately, suspensions may also dissipate
tensively in section 2, most arguments proposed by
power generated by the cyclist’s muscles through
manufacturers favouring their particular system still
small oscillatory suspension displacements often re-
need to be objectively assessed.
ferred to as ‘pogoing’
[2]
or, more commonly, ‘bob-
bing’.
[3,8,9,12]
Bobbing can essentially be generated
2. Mechanical Aspects: The Engineer’s
by two mechanisms: (i) the displacement of the
Point of View
cyclist’s body parts; and (ii) the interaction between
the forces applied on the pedals that are transmitted
By definition, the main function of suspension
to the front chain-ring and the rear suspension.
systems is to absorb energy. More specifically, the
The first mechanism can be reduced either by
viscous element (damper) present in most shock-
tuning the suspension optimally and/or by the cyclist
absorbing devices is precisely designed to dissipate
who can adapt his/her pedalling technique.
[2,10]
energy transmitted to the suspension element. In the
However, the magnitude of the energy dissipation
bicycle, cyclist and terrain model, energy can be
that may occur through such a mechanism has never
generated and transmitted to the suspensions either
been studied in detail. It seems clear that such an
by the terrain irregularities or by the cyclist them-
issue will be difficult to address since it may be
self
[8,9]
(figure 3).
assumed that displacement of a cyclist’s body parts
The aim of bicycle suspensions is to dissipate
is probably highly variable among individu-
energy generated by the terrain allowing better com-
als.
[14,16,17]
fort and bicycle handling for the cyclist. It is worth-
The second mechanism (bobbing induced by
interaction between front chain-ring and rear sus-
pension) is graphically represented in figure 4. As
recalled by Good and McPhee,
[13]
such a mechanism
may become exaggerated in high-load situations
such as climbing or sprinting. Several au-
thors
[3,8,12-14,17]
have studied this phenomenon in
more detail by modelling the bicycle and cyclist as a
multi-body system with springs and dampers in
order to evaluate the magnitude of energy dissipated
in the rear suspension.
More specifically, Wang and Hull
[11,12]
calculat-
ed that the power dissipated in the rear suspension
was 6.9W when cycling uphill at 6.5 m/sec (23.4
km/h) a 6% grade smooth surface. This 6.9W value
represents only 1.3% of the total power developed
Deformable cyclist
in movement
a
a
b
b
b
Rear
shock-absorbing
device
Front
shock-absorbing
device
Fig. 3. Forces transmitted to the suspensions. In off-road cycling,
forces transmitted to the suspensions may be classified in two
categories: (a) forces generated by the terrain irregularities; and (b)
forces generated by the movements of the cyclist which are applied
on the handlebar, saddle and pedals.
 2004 Adis Data Information BV. All rights reserved Sports Med 2004; 34 (2)

Bicycle Suspensions and Energy Expenditure 75
by the cyclist, which is in agreement with data
presented by Kyle.
[18]
Noteworthy, power dissipated
in the FS was found negligible. Wang and Hull
[11]
also validated their findings experimentally by mea-
suring front and rear suspension displacements with
linear transducers while a cyclist rode a commercial-
ly available DS bicycle on a treadmill at the same
grade and speed.
Later, Wang and Hull
[12]
studied the optimal rear
suspension pivot point location in terms of energy
loss minimisation. The vertical position of the pivot
point was the most critical factor. Their model
showed that power dissipated in the rear suspension
could be reduced to 1.2W when the pivot point was
positioned on the seat tube, 11cm above the bottom
bracket for a 32-teeth front chain-ring. They also
showed that optimal pivot point location was very
insensitive to the fore-aft location of the pivot point,
to pedalling mechanics, and to both spring and
damping parameters. Due to the U-shape of the
power dissipation versus vertical location of the
pivot point curve, a sub-optimal pivot point location
lead to only a small increase in energy loss. Wang
and Hull’s results also indicated that the optimal
pivot point location is directly dependent of the size
of the front chain-ring, which is understandable
because that parameter directly determines the posi-
tion of the chain-line.
Needle and Hull
[3]
conducted an experimental
study with a DS bicycle with adjustable geometry
and suspension parameters to verify the validity of
both the previously proposed model and the location
of the rear suspension pivot point obtained by simu-
lation. In this experimental study of suspension dis-
placements with only one cyclist, the optimal pivot
point location was found to be 8.4cm above the
bottom bracket, which was relatively close to the
result obtained by simulation.
Good and McPhee
[17]
developed a less complex
four-body dynamic model of a rear suspended bicy-
cle that was quite different than that of Wang and
Hull.
[11]
The response of their four-body model was
compared with simulation data previously obtained
by Wang and Hull.
[11]
Although this four-body
model was considerably simpler, it produced similar
 2004 Adis Data Information BV. All rights reserved Sports Med 2004; 34 (2)
a
Suspension
compresses
Pivot
Chain
force
b
Suspension
extends
Pivot
Chain
force
Fig. 4. Rear suspension compression (a) and extension (b) forces in relation to pivot point location. In bicycles equipped with rear suspensions, the cyclist-generated pedalling
forces are transmitted to the rear shock-absorbing devices. According to the location of the rear suspension pivot point, pedalling forces may cause the rear suspension to
compress (a) or to extend (b), which generates cyclist-induced power dissipation (adapted from Wang and Hull,
[12]
1997, with permission  Swets & Zeitlinger).

76 Nielens & Lejeune
results in terms of rear suspension displacements as (braking, cornering, traction etc.). However, in the
standing position, power dissipated in the rear sus-
a function of crank (bottom bracket) angle. Using a
pension may reach up to 5% of the total power
different optimisation method known as a ‘genetic
developed by some cyclists. It must be kept in mind
algorithm’ to determine the optimal design of the
that many simplifying assumptions are made in sim-
rear suspension (pivot point location) along with
ulation studies. Hence, numerous factors are ne-
their four-body model, Good and McPhee
[13]
show-
glected that may lead to a systematic underestima-
ed that the optimal rear suspension pivot point was
tion of power losses due to suspension systems, e.g.
located 11.6cm above the bottom bracket and 2.7cm
multi-articulated DS bicycle frames become more
behind the seat tube. Such results are again very
flexible. It may be hypothesised that the amount of
close to the first simulation data obtained by Wang
energy dissipated in the frame itself
[2]
may increase
and Hull.
significantly. Such a source of energy dissipation is
Finally, in a recent study by Karchin and Hull,
[14]
neglected in computer-simulation studies. However,
11 experienced cyclists were asked to ride a custom-
it may be significant as witnessed by the efforts
built DS bicycle with adjustable geometry and sus-
made by manufacturers of bicycle parts to reduce
pension parameters at an approximate power of
the flexibility of front telescopic-suspension forks
300W (6% grade on a treadmill at 24.8 km/h) in a
by designing reinforced front wheel hubs and brake
seated as well as in a standing position. By monitor-
bridges.
[1]
ing the suspension displacements with linear poten-
Another method to evaluate the amount of power
tiometers, Karchin and Hull showed that: (i) the
lost in suspensions is to measure the energy con-
minimum power loss at the optimal pivot point
sumed by the cyclist riding FS or DS bicycles as
height was quite variable among study participants
compared with non-suspended ones. Section 3 of
(mean 0.89W, range 0.59–1.25) indicating large
this manuscript will review such studies and focus
variability in pedalling mechanics; (ii) power dissi-
more on physiological variables that can be observ-
pated in the rear suspension was considerably higher
ed in the laboratory or on the field.
(mean 6.49W, range 0.7–13.48) in the standing posi-
tion; (iii) no significant interaction between front
3. Energetic Aspects: the Exercise
and rear suspensions could be found; (iv) the opti-
Physiologist’s Point of View
mal pivot point location is higher in the sitting
position (9.77cm) than in the standing position
The energy consumed by the cyclist ultimately
(5.88cm); and (v) notwithstanding the wide variabil-
reflects the result of the interaction of the multiple
ity in the minimum power loss among study partici-
variables from the terrain, bicycle and cyclist. On
pants, the optimal pivot point location remained
the same irregular terrain travelled at the same
consistent.
speed, a more efficient suspension system (i.e. a
In summary, all mechanical simulation and ex-
system that dissipates terrain-induced energy well
perimental studies specifically conducted to evalu- without dissipating cyclist-generated energy) should
allow the cyclist to expend less energy than with a
ate the power dissipated by suspensions in DS bi-
less efficient suspension, since less cyclist-generat-
cycles agree that the estimated power lost in the rear
ed power will be dissipated in the suspension. At
suspension ranges from 0.5–2W in the seated posi-
maximal energy expenditure rate, this more efficient
tion when the pivot point is located optimally
suspension will ultimately allow the cyclist to attain
(±10cm above the bottom bracket on the seat tube).
higher speed.
Such a magnitude of power loss remains inferior to
0.7% of the total power developed by the cyclist. It
Berry et al.
[4]
were the first to evaluate metabolic
should therefore be largely compensated by the ben-
expenditure by means of oxygen consumption
efits provided by suspensions in terms of comfort
(
˙
VO
2
) of cyclists riding different types of bicycles
and performance related to better bicycle handling (no suspension, FS, DS and rear suspension only) on
 2004 Adis Data Information BV. All rights reserved Sports Med 2004; 34 (2)

Bicycle Suspensions and Energy Expenditure 77
a treadmill with a 4% grade and at a speed of 10.4 significant 11.5% reduction in energy consumption
km/h with or without a 3.8cm high bump attached to observed in the DS mode by Berry et al.
[4]
Although
its belt with duct tape. No significant differences in up to 12 study participants took part in the first
˙
VO
2
were noted in relation to the suspension type in phase of the study where metabolic measurements
the no-bump condition. In other words, suspensions were conducted, a type II error may still not be
did not significantly increase energy expended by excluded. This hypothesis may be supported by the
the cyclist riding on a smooth surface. They observ- fact that mean heart rates recorded on the rigid
ed a very significant energy consumption increase in bicycle were significantly higher than those on the
relation to the presence of the bump, ranging from a suspended bicycles. In another phase of the proto-
63% increase compared with the no-bump condition col, seven study participants were asked to complete
for the non-suspended bicycle to 41% for DS bicy- three different time trial courses as fast as possible
cle. The DS allowed a very significant 11.5% de- including a downhill, a climb and a cross-country
crease in
˙
VO
2
(p = 0.004) compared with the non- course. In the cross-country trial, the best perform-
suspended condition. Adding only a rear suspension ance was achieved on the FS bicycle in five of the
to the bicycle already yielded a significant energy seven cyclists with the mean finishing time being
saving. Surprisingly, the FS alone did not succeed in significantly smaller (p = 0.02) with the FS. This
lowering energy expenditure significantly, which observation fits well with what is currently observed
suggests that FS does not succeed in reducing ener- in competition. No significant differences in finish-
gy consumed by the cyclist when a DS system does. ing times relative to suspension types were noted in
Such a finding contrasts with data of other authors the downhill and climbing trials.
that will be discussed further in the next paragraph.
More recently, MacRae et al.
[7]
conducted a study
However, a type II statistical error may not be
on two different outdoor uphill courses aiming at
excluded due to the relatively small number of study
comparing performances of six experienced cyclists
participants (n = 6) included in this study.
riding bicycles equipped with FS and DS. The first
Seifert et al.
[5]
were the first to evaluate the course was a 1.62km asphalted road with a 14.2%
energy cost of bicycle suspensions in a more natural mean grade. The second course was a ‘rocky and
outdoor environment. They designed a rather com- rutted’ fire access road that was 1.38km with a mean
plex protocol during which cyclists were asked to 11.3% grade. Traditional physiological variables
undergo three consecutive experimental phases us- (e.g. heart rate and
˙
VO
2
) were monitored and
ing different suspension types (no suspension, FS mechanical power generated by the cyclist was mea-
and DS). In one phase, 12 study participants rode at sured and recorded during the runs by a Schoberer
a constant speed (16.1 km/h) on a flat looped ~400m Rad Messtechnik (SRM) Training System (Weldorf,
course built on hard level ground with 45 fabricated Germany). On both courses, performances in terms
5cm high wooden bumps. It must be emphasised of finishing times were not significantly different
that such experimental conditions are relatively whether the study participants used the FS or the
close to those of Berry et al.
[4]
with 3.8cm high DS. Likewise, mean heart rate,
˙
VO
2
recorded during
bumps encountered 42 times/min and a 4% grade the runs and blood lactate concentration measured in
compared with 5cm high bumps encountered 30 samples collected 2 minutes after the end of the
times/min and no grade, respectively, in Berry et trials were not different, suggesting that metabolic
al.
[4]
and Seifert et al.
[5]
protocols. Only a trend cost is independent of the suspension types. How-
toward a significant reduction of mean
˙
VO
2
(p = ever, mean mechanical power generated by the
0.07) was observed with the suspended bicycles study participants as recorded by the SRM system
compared with the non-suspended one. Although was almost 30% higher with the DS compared with
experimental conditions were relatively similar, the the FS on both courses, which was very significant.
results from Seifert et al.
[5]
contrast with the very In our opinion, the findings of this study must be
 2004 Adis Data Information BV. All rights reserved Sports Med 2004; 34 (2)

78 Nielens & Lejeune
considered with caution. It is hard to explain how suspension modes in relation to the relatively high
deformability of such articulated frames and forks.
the study participants could have developed close to
The results of this study should therefore not be
30% more mechanical power on the DS bicycle with
extended to compare high-level cross-country com-
all physiologically recorded parameters remaining
petitors using traditional telescopic FS and DS with-
unchanged and when they were all asked to perform
out caution.
maximally on both bicycle types on both courses.
In summary, Berry et al.
[4]
were the first to report
The authors argue that the usual relationship be-
laboratory data suggesting that the metabolic cost of
tween power output, cardiovascular and metabolic
riding a bicycle on a bumpy surface could be very
responses usually observed in the laboratory might
significantly reduced by DS systems. Such observa-
be altered on the field. However, it must be noted
tion illustrates the fact that terrain-induced energy
that the SRM power-meter system has not been
can indeed be dissipated in the suspension system,
validated when used with DS bicycles. Another
allowing the cyclist to consume less energy to travel
hypothesis could be that interactions between the
at the same speed over bumpy and shaky surfaces.
front chain-ring (on which the SRM system is
The fact that suspensions may also dissipate the
mounted on) and the rear suspension could have led
cyclist-generated power remains a major concern
to a systematic mechanical power overestimation
for bicycle manufacturers and competitors. The only
due to so-called ‘kick-back’ effects.
[12]
three studies
[4,6,7]
that evaluated metabolic cost
Finally, Nielens and Lejeune
[6]
studied the meta-
(
˙
VO
2
) of cyclists using suspended bicycles on
bolic cost of riding a bicycle equipped with different
smooth surfaces failed to demonstrate any increase
types of suspensions on a smooth surface. In this
in metabolic cost in relation to suspensions. Unfor-
experiment conducted in a laboratory, one cross-
tunately, the conclusions that can be drawn at this
country DS bicycle was mounted on a bicycle train-
point are limited because of the paucity of studies
er. Different suspension systems were obtained by
that evaluated metabolic cost on the field, with the
successively replacing rear and both suspension ele-
rather poor statistical power of some studies and,
ments by rigid links. The aim of the study was to
finally, with the relatively low value of the mechani-
specifically evaluate any possible cyclist-induced
cal power outputs that have been investigated.
energy loss when riding a modern cross-country
suspended bicycle after eliminating terrain-induced
4. Bicycle Comfort and Physical Stress in
energy losses by riding on a smooth surface. Twelve
Relation to Suspensions
study participants were asked to perform a 15-min-
ute gradational cycling protocol starting at 50W
Seifert et al.
[5]
evaluated perceived riding com-
with 50W increments every 3 minutes in the three
fort and rate of perceived exertion of 20 cyclists who
suspension modes. No difference in
˙
VO
2
, nor in
successively rode non-suspended, FS and DS bi-
heart rate in any of the stages of the tests was
cycles on a hard, level ground course with 45
observed in relation to the suspension type, sug-
fabricated bumps during approximately 1 hour. The
gesting that suspensions do not generate any extra
DS bicycle was perceived as the most comfortable,
energy expenditure for the cyclist when riding on a
and the FS as more comfortable versus the non-
smooth surface. However, in this experiment, the
suspended bicycle. Perceived exertion data favoured
highest external power reached was 250W, which is
the suspended bicycles and no significant difference
probably significantly lower than powers attained
was observed between suspension types. In the same
by competitors in actual cross-country races. More-
study, Seifert et al.
[5]
also reported lower creatine
over, in all suspension modes, the bicycle frame
kinase levels in the venous blood samples of cyclists
remained the same (a modern light cross-country
after riding the suspended bicycles (FS and DS)
frame with four-bar linkage front and rear suspen-
compared with rigid ones. Although no significant
sions) and energy losses may have occurred in all
difference was observed between DS and FS for
 2004 Adis Data Information BV. All rights reserved Sports Med 2004; 34 (2)

Bicycle Suspensions and Energy Expenditure 79
creatine kinase change, these data strongly suggest gesting that suspensions dissipate terrain-induced
that suspension systems are effective in reducing energy variations well. Two studies
[4,6]
conducted in
muscular stress. a laboratory reported no energy expenditure in-
crease related to suspensions when riding on a
In summary, comfort is obviously an important
smooth surface. Such studies indicate that if any
issue for recreational cyclists who often favour soft-
energy was indeed dissipated in the suspensions, it
er and more comfortable suspension systems.
was too small to be measured by traditional respira-
Nevertheless, the fact that suspensions seem effec-
tory gas analysis methods. In other words, the mag-
tive in reducing physical stress may become partic-
nitude of any energy dissipation by modern suspen-
ularly relevant for competitors engaged in long dis-
sion systems must be very small, if any, and thus
tance events that may last up to 6–12 hours and for
probably negligible compared with the advantages
recreational cyclists who generally favour comfort.
they provide. Such results are in agreement with
data observed in mechanical simulation studies.
5. Conclusion
In summary, evidence present in the literature
Bicycle suspensions have been designed to im-
suggests that cyclist-generated power that is dissi-
prove bicycle comfort and handling by dissipating
pated by suspensions is minimal and probably negli-
terrain-induced energy variations. However, they
gible on most terrains. However, the scarce studies
may also dissipate the energy generated by the cy-
on the topic, as well as the limitations in the conclu-
clist through small oscillatory movements often
sions that can be drawn from most of them, indicate
termed ‘bobbing’. Bobbing is generated by the cy-
that we should remain careful before supporting the
clist’s body movements as well as by the forces
use of DS bicycles on all course-types and for all
exerted on the pedals that interact with the rear
cyclists. For instance, the fact that most cross-coun-
suspension. This phenomenon is a major concern for
try racers still use FS bicycles should be kept in
bicycle manufacturers and competitive cyclists en-
mind.
gaged in events where most of the time is spent
Finally, suspension systems clearly improve
climbing, such as cross-country races. Any cyclist-
comfort and reduce physical stress. This issue
generated power dissipated in the suspensions will
should not be overlooked, as most recreational cy-
slow down the cyclist. Ideally, such a loss in per-
clists will ultimately favour comfort over perform-
formance that mainly takes place when riding uphill
ance.
should always remain smaller than the gain in speed
provided by suspensions through better bicycle han-
Acknowledgements
dling, mainly in the downhills.
The authors wish to thank Mr E.J. Conley for his valuable
Several mechanical simulation studies conducted
help in the writing of this manuscript. No sources of funding
by engineers suggest that if bicycle suspensions and
were used to assist in the preparation of this manuscript. The
geometry are optimised, cyclist-generated power
authors have no conflicts of interest that are directly relevant
that is dissipated when riding in a seated position is
to the content of this manuscript.
minimal if not negligible. However, several possible
dissipation sources are neglected in such studies as
References
they included numerous necessary simplifying as-
1. De Lorenzo DS, Wang EL, Hull ML. Quantifying off-road
sumptions in the models.
suspension hub stiffness. Cycling Sci 1994; 3: 12-26
2. Olsen J. Bicycle suspension systems. In: Burke ER, editor.
Only a few studies have evaluated the effects of
High-tech cycling. Champaign (IL): Human Kinetics, 1996:
suspension systems on the energy expended by the
45-64
cyclist in the laboratory and in the field. One study
[4]
3. Needle SA, Hull ML. An off-road bicycle with adjustable
suspension kinematics. Cycling Sci 1997; 1: 4-29
showed that the energy expended by the cyclist
4. Berry MJ, Woodard CM, Dunn CJ, et al. The effects of a
could be very significantly reduced by a DS system
mountain bike suspension system on metabolic energy expen-
when riding on an artificial bumpy course, sug- diture. Cycling Sci 1993; 3: 8-14
 2004 Adis Data Information BV. All rights reserved Sports Med 2004; 34 (2)

80 Nielens & Lejeune
5. Seifert JG, Luetkemeier MJ, Spencer MK, et al. The effects of 13. Good C, McPhee J. Dynamics of mountain bicycles with rear
mountain bike suspension systems on energy expenditure, suspensions: design optimization. Sports Eng 2000; 3: 49-55
physical exertion, and time trial performance during mountain
14. Karchin A, Hull ML. Experimental optimization of pivot point
bicycling. Int J Sports Med 1997; 18: 197-200
height for swing-arm type rear suspensions in off-road bi-
6. Nielens H, Lejeune TM. Energy cost of riding bicycles with
cycles. J Biomech Eng 2002; 124: 101-6
shock absorption systems on a flat surface. Int J Sports Med
15. Whitt F, Wilson D. The wheel: bicycling science. Cambridge
2001; 22: 400-4
(MA): The MIT Press, 1989
7. MacRae H-H, Hise KJ, Allen PJ. Effects of front and dual
16. Daly DJ, Cavanagh PR. Asymmetry in bicycle ergometer pedal-
suspension mountain bike systems on uphill cycling perform-
ling. Med Sci Sports 1976; 8: 204-8
ance. Med Sci Sports Exerc 2000; 32: 1276-80
17. Good C, McPhee J. Dynamics of mountain bicycles with rear
8. Wang EL, Hull ML. A dynamic system model of an off-road
suspensions: modeling and simulations. Sports Eng 1999; 2:
cyclist. J Biomech Eng 1997; 119: 248-53
129-43
9. Wang EL, Hull ML. Power dissipated by off-road bicycle
18. Kyle C. Chain friction, windy hills, and other quick calculations.
suspension systems. Cycling Sci 1994; 4: 10-26
Cycling Sci 1990; 2: 23-6
10. Olsen J. Bicycle suspension meet Mr Simple Dynamics. Cycling
Sci 1992; 3: 6-12
11. Wang EL, Hull ML. A model for determining rider induced
Correspondence and offprints: Dr Henri Nielens, Sports
energy losses in bicycle suspension systems. Vehicle Syst
Medicine, Cliniques universitaires Saint-Luc, Universit
´
e
Dynam 1996; 25: 223-46
catholique de Louvain, Avenue Hippocrate, 10, 1200 Brus-
12. Wang EL, Hull ML. Minimization of pedaling induced energy
sels, Belgium.
losses in off-road bicycle rear suspension systems. Vehicle
Syst Dynam 1997; 28: 291-306
E-mail: nielens@read.ucl.ac.be
 2004 Adis Data Information BV. All rights reserved Sports Med 2004; 34 (2)